SEARCH

SEARCH BY CITATION

Keywords:

  • Passive sampling;
  • Monitoring;
  • Emerging contaminant;
  • Bioassay

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

The implementation of strict environmental quality standards for polar organic priority pollutants poses a challenge for monitoring programs. The polar organic chemical integrative sampler (POCIS) may help to address the challenge of measuring low and fluctuating trace concentrations of such organic contaminants, offering significant advantages over traditional sampling. In the present review, the authors evaluate POCIS calibration methods and factors affecting sampling rates together with reported environmental applications. Over 300 compounds have been shown to accumulate in POCIS, including pesticides, pharmaceuticals, hormones, and industrial chemicals. Polar organic chemical integrative sampler extracts have been used for both chemical and biological analyses. Several different calibration methods have been described, which makes it difficult to directly compare sampling rates. In addition, despite the fact that some attempts to correlate sampling rates with the properties of target compounds such as log KOW have been met with varying success, an overall model that can predict uptake is lacking. Furthermore, temperature, water flow rates, salinity, pH, and fouling have all been shown to affect uptake; however, there is currently no robust method available for adjusting for these differences. Overall, POCIS has been applied to a wide range of sampling environments and scenarios and has been proven to be a useful screening tool. However, based on the existing literature, a more mechanistic approach is required to increase understanding and thus improve the quantitative nature of the measurements. Environ. Toxicol. Chem. 2012; 31: 2724–2738. © 2012 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

The need for polar passive samplers

Within environmental science there is an increasing focus toward so-called emerging contaminants. Many of these compounds are polar or semipolar, such as pharmaceuticals and personal care products (PPCPs), and thus, their behavior and fate in the environment can be very different from more traditionally studied persistent organic pollutants. These compounds are just beginning to be included in legislation, for example, in the amended list of priority substances for the European Water Framework Directive. Due to their potency at low concentrations, several of these additions to the priority list have environmental quality standard values set at extremely low levels (e.g., 0.035 ng/L for the pharmaceutical 17α-ethinylestradiol [EE2] and 0.65 ng/L for the industrial chemical perfluorooctane sulfonic acid) 1. The measurement of polar compounds in environmental matrices and at such trace concentrations represents a significant challenge and has become possible only in recent years due to significant improvements in analytical techniques. In this regard, passive sampling devices (PSDs) may have much to offer the analytical process by providing a time-integrated sample with low detection limits and in situ extraction of analytes (http://www.norman-network.net/public_docs/slides_prague/norman_position_paper_pas_sampling.pdf). Such PSDs are fairly well developed for hydrophobic compounds such as polychlorinated biphenyls, for example, the semipermeable membrane device (SPMD), low-density polyethylene (LDPE), and silicone rubbers. However, although samplers suitable for polar compounds have been available for some years, they remain more poorly characterized in terms of modeling uptake rates and the effects of environmental factors. In the present study, we provide a critical review of the calibration and use of the most popular of these samplers, the polar organic chemical integrative sampler (POCIS).

Polar organic chemical integrative sampler

The polar organic chemical integrative sampler consists of a receiving phase (sorbent) sandwiched between two polyethersulfone (PES) microporous (usually with 0.1 µm pore size) membranes 2. The sampler is generally compressed together using two stainless steel rings (interior diameter usually 51–54 mm), which expose a surface area of 41 to 46 cm2. In the original description of POCIS the exposed membrane was 3.3 cm in diameter, giving a surface area of 17 cm2 and a recommended membrane to sorbent ratio of 180 cm2/g 2. The two commercial versions are POCIS “pharmaceuticals,” which contain the widely used Oasis hydrophilic–lipophilic balance (HLB) sorbent, and POCIS “pesticides,” which contain a triphasic sorbent admixture 80:20 (wt:wt) of hydroxylated polystyrene-divinylbenzene resin (Isolute ENV+) and a carbonaceous sorbent (Ambersorb 1500), dispersed on S-X3 biobeads. This second configuration provided superior uptake and recovery for certain classes of polar compounds including pesticides and hormones 2. Several deviations from these standard configurations are reported, including sizes, types, and amounts of sorbents, membranes, and so forth, which are presented in the following text where appropriate.

Basic POCIS theory and modeling

A comprehensive analysis of relevant uptake theory and modeling applicable to POCIS is not the purpose of the present review and is covered in detail elsewhere 2, 3. In addition, comprehensive descriptions of general passive sampling theory, including different types of calibration experiments, are also provided elsewhere 4, 5. Thus, we limit the following section to several fundamental issues that facilitate an understanding of the subsequent text.

Accumulation in the sampler is driven by a difference in chemical activity of the substance dissolved in water and in the receiving phase of the sampler, initially free of the contaminant of interest. Accumulation of polar contaminants into POCIS devices is the result of successive processes occurring at the surface of the membrane and within the device. First, analytes dissolved in water have to cross (by diffusion) the water boundary layer (WBL), a retarded layer of water that develops due to friction and results in increased viscosity at the surface of the sampler 6. The thickness of this boundary layer is dependent on water flow/turbulence around the sampler and can affect accumulation rates significantly as observed for samplers for nonpolar substances 6, 7. Second, analyte transport across the PES membrane can be through water-filled pores of the microporous PES membrane or via the polymer itself. The average thickness of the hydrated PES membrane is 130 µm, while the estimated water-filled pore volume is 76.5% of the total membrane volume 3. Finally, compounds transfer from the PES membranes to the sorbent material mainly through adsorption. These last two steps render the understanding, modeling, and prediction of accumulation of the wide range of possible chemicals by POCIS challenging. As opposed to the absorption process that governs nonpolar contaminant accumulation, adsorption is a concentration-dependent phenomenon in nonporous polymeric samplers 5. Sorption isotherm experiments are needed to understand the distribution of polar substances between sorbent and water over a range of concentrations. These can be undertaken for individual compounds or as mixtures because competition between analytes for the same sorption sites on the sorbent is possible. For extended exposures, biofouling of the surface of the membrane needs to be accounted for.

Though they are assumed to be additives, very little is known for any chemical of the relative resistances to mass transfer in each of these compartments. In addition, there may be WBLs on the inside of the POCIS associated with both the membrane and the sorbent particles 3. For nonporous sorbents, analytes will usually bind with the active site at a rate which is too fast for this step to become limiting. For porous sorbents (such as Oasis HLB), analytes have to diffuse first through the pores to reach the binding site, although considering the other resistances potentially influencing uptake, such as multiple WBLs, it seems unlikely that this is the rate-limiting step. Boundary-layer control of the uptake in POCIS is generally exemplified by increases in observed sampling rates with increases in water turbulence 2, 3. When water turbulences are high enough for the resistance to mass transfer in the boundary layer to be negligible, transport across the PES membrane dominates the overall resistance and becomes the rate-limiting factor in contaminant uptake by POCIS. However, a comprehensive determination of the range of water turbulences that will affect sampling rates significantly is lacking. Experiments dedicated solely to the measurement of analyte mass transfer coefficients in the membrane are yet to be conducted in detail, but some water–membrane partitioning coefficients are provided by Vermeirssen et al. 8. In that study, compounds with log KOW > 2 clearly showed a lag phase in uptake due to accumulation in the PES membrane. The importance of the PES membrane is also exemplified for Empore disk–based samplers, where its addition slows sampling rates considerably compared with using naked discs, suggesting the introduction of a degree of membrane control into the uptake process 9–12. Detailed information on the effects of environmental parameters on sampling rates is also required to obtain quantitative data.

Previous reviews

Several reviews cover the use of PSDs for environmental monitoring in general 13–15 and in the aquatic environment 16, 17, or consider specific types of hydrophobic passive sampler such as SPMDs 18, 19. Additionally, polar passive samplers (including POCIS) are mentioned as part of several reviews considering the measurement of specific groups of compounds such as PPCPs 20–22, emerging polar contaminants in general 23, 24, or in relation to wastewater 25. The focus of the present review is on critically evaluating several aspects of the use of POCIS, particularly its calibration and the effects of environmental conditions on the uptake process.

CALIBRATION METHODS USED FOR POCIS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Rationale for calibration

Passive sampler calibration generally involves exposing samplers to water spiked with a known concentration of analytes of interest under controlled conditions in the laboratory 5, 16. This way a sampling rate (Rs) can be estimated, which is the equivalent volume of water extracted by the sampler per unit of time (typically given as liters per day). These experiments can also be designed to establish exposure times during which contaminant accumulation remains in the linear phase of uptake or whether contaminant uptake is concentration-independent (prerequisites for quantitative use). An understanding of the effects of environmental variables on POCIS sampling rates is required to ensure quantitative information can be obtained from field deployments 16. Calibration can, however, also be synonymous with experiments specifically designed to understand processes governing solute–sampler interactions. For passive samplers of nonpolar organic pollutants, these have focused on the measurement of polymer–water partition coefficients 26, 27 or of contaminant diffusion coefficients in polymers 28, 29. These supporting data often help to explain the mechanisms of analyte uptake into passive samplers and experimental sampling rates and their modeling 4. Over the last decade, numerous POCIS calibration studies have been conducted to derive Rs estimates for herbicides such as triazines, polar pesticides such as phenyl ureas, a wide range of PPCPs, endocrine-disrupting chemicals (EDCs), natural/synthetic hormones, and illicit drugs (see Reported sampling rates and Reported Applications and Use below). Different procedures varying in their level of complexity and resources needed have been applied to determine POCIS sampling rates 2, 30, 31.

Static renewal

The first POCIS calibration experiments were undertaken using a static renewal design, where samplers were exposed to a relatively small volume of exposure water (1–3 L) refreshed periodically 2. The POCIS was then removed at intervals, and the mass of analytes sorbed was measured in POCIS extracts. Such a procedure has been used in several instances since 32–35. The frequency for refreshing the test water should be high enough to minimize the decrease (ideally <10% per refreshment period) in analyte concentration caused by uptake into the sampler(s) during each batch exposure and allows for a more straightforward modeling of sampling rates. Ideally, the exposure concentration should be measured at least at the start and end of each batch renewal 5. Often, daily renewals are necessary, which results in an intensive sampling of the water phase for calibrations typically running for up to one month. Despite this, static renewal procedures are convenient. They are generally conducted in 1- to 3-L beakers, although volumes of up to 80 L are also reported 36; and replicate exposures can easily be prepared using different types of water. Larger tank volumes may appear to allow a static calibration where exposure water depletion is negligible and, thus, renewal not needed 30, although careful attention must still be paid to the water concentrations and additional spikes may be required 36.

Static depletion

The design of static depletion experiments revolves around a single spike of a chemical to the water, following which the drop in analyte concentration in the exposure solution, often covering some orders of magnitude, is monitored. Temporal changes in analyte concentration in water are used to estimate Rs 37–42. First-order dissipation kinetics are in general subsequently fitted to decreases in solute concentration in water observed with time. High initial analyte concentrations are often used (i.e., >5–10 µg L−1) because they allow the decrease in analyte concentration in the exposure solution to be followed over three to four orders of magnitude 37. Positive controls are often needed for such calibrations to assess analyte losses due to processes other than sampler accumulation (e.g., volatilization, photodegradation, or hydrolysis). For example, a short half-life for omeprazole (antiulcerative) of 2 d was observed in POCIS calibration experiments by MacLeod et al. 38, and a significant decrease in concentration of ketoprofen was observed over 50 h in experiments by Togola and Budzinski 43. These authors noted the importance of relying on analyte masses sorbed to the POCIS sorbent for Rs estimation rather than only the drop in analyte concentration in water. The importance of using adequate positive controls in these calibrations should not be underestimated, especially in the light of evidence that a significant amount of some analytes may be contained within the membrane 8, 44, which is usually not analyzed or included in the calculations. Despite this, some authors do not appear to have conducted any such control 42 or describe them as “fortified water” 40 or exposure water with either “no POCIS” 38 or just the steel rings without any membranes 41. Static depletion designs have proved useful to investigate the effects of dissolved organic matter (DOM) or pH on Rs, for example Vrana et al. 39, covered in more detail below (Factors Affecting POCIS Sampling Rates).

Flow-through systems

Flow-through experiments are designed with the aim of maintaining a constant analyte concentration in water. This is achieved with a continuous supply of fortified water to the exposure tank where multiple samplers are deployed under identical hydrodynamic conditions 45. Constant concentrations can be obtained if the flushing rate is able to compensate for analyte losses and removal from solution by the samplers 5. Importantly, all samplers are in principle exposed to the exact same analyte concentration in water in this experimental design. Samplers are removed from the exposure tank over time, and the sorbent is analyzed to assess the uptake of analytes by POCIS. Relatively high exposure concentrations are reported (e.g., 1 µg/L) 2 for static/static renewal exposures as this enables the collection of small enough aliquots of exposure solution without influencing the test. However, for flow-through exposures, lower exposure concentrations (<100 ng/L) can be used because the collection of larger water samples will not affect the calibration as spiked water is being added continuously. Analytes in acetone or methanol can be injected directly into the calibration tank or mixed in water before it is pumped into the calibration tank. In addition, it is worth noting that flow-through systems avoid the positive control requirements of the static experiments. Flow-through calibrations have been undertaken for alkylphenols 46, 47 and EDCs 48, for example.

In situ calibration

In situ calibration is a procedure that can be used to measure sampling rates in the field at the exact location where measurements are to later take place and has successfully been carried out for hydrophobic PSDs 49. Samplers deployed in the field are calibrated by comparing analyte accumulation with time-averaged concentrations in water obtained from high-frequency grab sampling 31. While simple models can be used to calculate Rs in cases where analyte concentrations are relatively constant, models also exist for when analyte concentrations in exposure water are more variable 5. In a recent study by Harman et al. 31 an in situ calibration of POCIS devices was undertaken in a sedimentation overflow channel of a sewage-treatment plant. The channel provided a system with relatively stable conditions of flow and temperature. Samplers were exposed for periods of 3, 7, 14, 21, and 31 d over a five-week-long calibration during which high-resolution automatic sampling (6-h time proportional samples) was conducted for a range of illicit drugs. Sampling rates were then used to interpret data for a year-long consecutive 14-d POCIS exposure 31.

In the absence of a performance reference compound (PRC)–based system (see below, Development of a PRC approach for POCIS), which can link lab-based Rs to those occurring in field deployments, in situ calibration is the technique that will provide the best possible approximation of time-weighted average concentrations from POCIS deployments, if this is of great importance to the study. This is due to deployment factors such as the chemical composition of the matrix, pH, dissolved organic carbon, competitive sorption, temperature, and flow rates, likely being more similar to those during actual field deployments compared to those under laboratory calibration conditions. This appears to be confirmed by Zhang et al. 48, who found much higher sampling rates for a suite of EDCs when calibrated in situ compared to laboratory-derived values. Although a considerable amount of effort is required to perform such calibrations, which may appear prohibitory, the need for positive controls and maintaining an experimental laboratory setup is removed.

Outlook for POCIS calibrations

Which calibration method is most appropriate depends on the question to be answered. Thus, in situ calibration should be considered where the accuracy of estimated water concentrations is imperative or when sampling is carried out over long time periods. Conversely, where a simple, cost-effective estimation of sampling rates is required, static renewal experiments might be sufficient. In the absence of any evidence in the literature to the contrary, it appears convenient to standardize static renewal calibrations to that originally described by Alvarez et al. 2. A summary of the advantages and disadvantages of the various POCIS calibration methods is given in Table 1.

Table 1. Main characteristics, advantages, and disadvantages of different polar organic chemical integrative sampler calibration procedures available (based on a typical four-week calibration)
MethodMain characteristicsAdvantagesDisadvantages
Static depletion1- to 3-L exposures in beakersSimple procedureInability to systematically test the impact of water turbulences
 High initial solute concentration in waterMany replicates can be preparedInability to evaluate concentration dependence of sampling rates
 Measurement of solute concentration in water (at least five or more time points)Ability to test many types of waters (e.g., tap water, seawater, freshwater)Unrealistically high starting exposure concentrations
 Rs calculation: First-order dissipation model fitted to changes in concentration in waterAbility to vary certain environmental variablesInability to assess linearity of solute accumulation with time
  Small number of samples that require chemical analysisIssues with positive controls
  Low cost 
Static renewal1- to 3-L exposures in beakersSimple procedureIntensive exposure water sampling and analysis
 Daily–several days renewal frequency of exposure solutionMany replicates can be preparedRelatively complex Rs calculation when exposure concentration drops significantly
 Measurement of solute concentration in water at least as often as renewed and accumulation in samplers (at least four time points)Ability to test many types of waters (e.g., tap water, seawater, freshwater)Inability to systematically test the impact of water turbulences
  Ability to vary certain environmental variablesLabor-intensive
Flow-through>20-L exposure tankPossibility to systematically test effects of a wide range of environmental variablesBiofouling is possible for long exposures (e.g., one month)
 Constant flow-through of fortified exposure solutionAbility to obtain constant exposure concentrationsDifficulty in setting up a system in a temperature-controlled room
 Measurement and analysis of exposure solution and of samplers (four or more time points)Collection of water samples is simpleRelatively large volumes of water and amount of solute needed
  Long-term calibrations are possibleGeneration of large amounts of wastewater
  Many samplers exposed to identical water and solute concentrationsExpensive
In situ calibrationIn situ exposureSimpleRequires very intensive/high frequency water measurements to account for fluctuations (Autosampler may be used)
 Dedicated on-site sampling cells or channelsSampling rates are obtained for conditions that are very similar to future field deploymentsCompounds need to be in the exposure medium without spiking
 Measurement and analysis of exposure solution and of samplersNo positive controls requiredChallenge with the measurement of the truly dissolved fraction
 Sampler deployment in an identical housing, and position to when measurement is neededFocus on relevant compoundsPotentially complex modeling
 Rs calculation: From the kinetics of solute accumulation in the samplers and time-weighted average exposure solution concentration  

It is desirable that future calibration studies focus on a more systematic testing of changes in sampling rates with changes in important environmental variables. While a large number of calibration experiments have been conducted using static exposure designs, it is important that the sampling rates generated can be related to those from other designs. A first step in this process may be to include widely studied compounds in all calibration experiments where possible (suitable compounds suggested below in Reported Applications and Use) as well as some interlaboratory studies to determine variability between calibrations. Future calibrations should also focus on gaining a more detailed functional understanding of solute–sorbent interactions for POCIS, along the lines of recent work by Bauerlein et al. 50, who examined which function groups govern sorption for Oasis HLB. This will help in developing models to predict which compounds will be retained on POCIS sorbents and hopefully reduce the use of resource-intensive calibration experiments for this purpose (solid phase extraction–type method development may also be used instead, see below in Types of POCIS used).

The processes of sorption and desorption that have been observed in several studies 30, 31, 47, 51 and their dependence on exposure concentration or the presence of multiple substances competing for the same sorption sites need to be examined. In addition, the water concentrations of many compounds that are sampled by POCIS can vary over orders of magnitude very rapidly, for example, in wastewater 52. Thus, understanding sampler reaction to varying solute concentrations in water is vital if an accurate interpretation of masses accumulated in the samplers is to be achieved. While modeling of sampler reaction to varying concentrations has been undertaken for nonpolar substances 53–55 and for medium polar herbicides using SDB-RPS Empore disks 56, limited data are available for POCIS 57. Calibration experiments simulating variable water concentrations should therefore be undertaken to develop and validate models that can describe how accurately POCIS integrates peaks in analyte concentrations.

FACTORS AFFECTING POCIS SAMPLING RATES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Water flow rates

At least five studies have compared POCIS sampling rates under conditions of differing water turbulences (Table 2). These have largely been carried out using static- or static renewal–type calibrations where the exposure water is still or stirred using a magnetic stirrer 2, 34, 40 or stirred at different rates 38. All these studies report severalfold increases in Rs (Table 2). However, making generalizations about these reported effects is difficult due to other factors also being different between studies (e.g., temperature) and the fact that flow rates are seldom measured with any accuracy and/or that they may represent poorly what is actually happening at the sampler surface. For example, the flow rate of 4.5 m/s in a 2-L beaker stated by Bartelt-Hunt et al. 37 seems highly unlikely to provide the same flow regime as 4.5 m/s in a river. Thus, while such experiments are simple to perform, they are not representative of conditions found in the environment. However, Charlestra et al. 58 recently showed similar uptake of pesticides when comparing stirred (magnetic stirrer at 40 rpm in a 1-L exposure vessel) batch experiments with those gained from a flow-through system (flow ∼0.32 cm/s). Both showed higher uptake (generally <×2) than quiescent experiments that were also carried out. In a more comprehensive study, Li et al. 59 studied uptake of PPCPs from municipal wastewater at four different flow rates (2.6, 5.5, 15, and 37 cm/s) using a system of channels 9. Sampling rates were not calculated, but the authors describe a less than twofold increase in accumulated amounts between 2.6 and 37 cm/s, with a few exceptions (e.g., trimethoprim and triclosan, 3.5- and sevenfold increases, respectively). These generally small differences (<×2, Table 2) led those authors to conclude that flow correction may not be necessary in studies where water flow rates vary over the range studied. Similar flow effects were observed when using a PES membrane but with an Empore disk receiving phase 60.

Table 2. Reported effects of environmental factors on polar organic chemical integrative sampler (POCIS) sampling rates (irrespective of POCIS configuration)
FactorRangeEffects on RsTarget compounds (n)References
  • a

    No replication.

  • b

    As nanograms POCIS−1, not sampling rate.

  • c

    Static condition also “slightly stirred.”

  • d

    Estimated flow rates.

  • e

    Reported as total organic carbon concentrations.

  • PCP = personal care product; EDC = endocrine-disrupting chemical; EE2 = 17α-ethinylestradiol; DOM = dissolved organic matter; NOM = natural organic matter; PSU = practical salinity units; BPA = Bisphenol A.

Temperature10–20°CIncrease <×1.5Hormone (1)41
 15–21°CIncrease ×1–2aPharmaceuticals (12)43
 5–25°CIncrease <×2 (generally)Pharmaceuticals/PCPs/EDCs (29)42
Water turbulenceStatic-stirredIncrease ×4–9Pharmaceuticals/pesticides (6)2
 Static-stirredIncrease ×4–5Natural toxins (2)34
 2.6–37 cm s−1Increase <×2b (generally)Pharmaceuticals/PCPs/EDCs (24)59
 Static-stirredcIncrease <×3 (generally)Pharmaceuticals/PCPs/EDCs (29)40
 3–12 cm s−1dIncrease <×7 (generally)Pharmaceuticals/PCPs (25)38
Fouling0.2–1.5 vs 0.5–2.4 g dm2−1Increase <×1.5 (generally)Alkylphenols (21)47
pH3.7–9Dependent on species (increase <×3 with increasing pH for basic compounds and decreased for acidic compounds)Pharmaceuticals/PCPs/EDCs (21)39
 4–10SimilarEE2 BPA + 2 not given (4)48
Salinity0–35 PSUDecrease similar <×0.4Pharmaceuticals (13)43
 0–35 PSUSimilarNot given (4)48
DOM3–5 mg/LNo significant effectPharmaceuticals/PCPs/EDCs (21)39
NOMe0.5–5 mg/LNo significant effectPesticides (5)58

An increase in sampling rates due to an increase in water flow rates or turbulences is due to WBL control of uptake in hydrophobic passive samplers 6, 7, 45, 55, 61. As similar observations are made for POCIS, WBL control of uptake has been assumed for POCIS as well. Thus, it is also assumed that eventually the WBL will be reduced to a thickness at which further increasing turbulences will no longer affect Rs as the membrane matrix and/or water-filled pores will dominate the overall uptake resistance 3. From the literature available it is apparent that increases in water flow rates increase POCIS sampling rates generally by a factor of <2. However, while this suggests WBL control of uptake, the situation may be more complicated for POCIS than for hydrophobic samplers. It is interesting to note that the effect appears to be compound-specific, for example, largest for more hydrophobic compounds in the study by Li et al. 59. This may be due to hydrophobic compounds being hindered more by the WBL than polar ones and due to increasing affinity for the PES membrane with hydrophobicity 8. In such cases, the question arises as to what extent WBL controls uptake into the membrane or whether the membrane itself influences accumulations within the sorbent. In conclusion, future work should focus on elucidating the uptake pathways taken by individual compounds and to what extent these may be affected differently by changes in exposure conditions.

Temperature

Several studies have examined the effects of water temperature on POCIS sampling rates (Table 2). In a comprehensive analysis, Li et al. 40 found relatively small differences in sampling rates with increasing temperature (twofold increase or less, from 5 to 25°C) for a range of nearly 30 PPCPs and EDCs. This is similar to the results of Togola and Budzinski 43, who found an increase in Rs of up to twofold (and a corresponding difference in estimated water concentrations) for most compounds over a temperature range of 15 to 21°C. Carbamazepine was an exception, where the increase of 6% was within the variation of the data. These results coincide closely with the twofold increase in diffusion coefficients over a temperature range of 20°C, which was predicted to cause a 50% increase in Rs by Alvarez et al. 2.

Fouling

It is generally stated that the PES membrane used in POCIS is less susceptible to fouling than those used in hydrophobic samplers such as SPMDs 2, 3. Polyethersulfone is, however, not immune to fouling, the level of fouling instead being dependent on the exposure environment (Fig. 1). In a comparison of fouling for SPMDs and POCIS, the polar sampler actually showed more fouling overall (0.2–2.8 vs 0.1–1.4 g dry wt/dm2 membrane area) 47. However, it is difficult to know how such results from controlled laboratory experiments relate to actual environmental exposures. A further interesting finding in that study was that sampling rates actually increased up to 55% for some alkylphenols in fouled compared to nonfouled samplers.

thumbnail image

Figure 1. Differences in polar organic chemical integrative samplers (POCIS) fouling from different deployment environments. Very lightly fouled POCIS (upper picture) deployed in the Norwegian Sea (∼200 km offshore) for 42 d and moderately fouled POCIS (bottom picture) from wastewater treatment works after 14 d exposure. [Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com.]

Download figure to PowerPoint

For hydrophobic samplers, the problem of a reduction in sampling rates due to fouling is assumed to be corrected by using the PRC approach, and several studies show both reduced dissipation of PRCs and uptake of target compounds due to fouling 47, 62, 63. It is likely that the effect of fouling on the sampling rates of POCIS will be compound-dependent, which may simply represent differences in the affinity of different target compounds to the fouling layer. As the fouling layer will largely consist of organic material, it will be reasonable to assume that the effect will be more significant for less polar compounds, although there is no overall trend toward that effect in the three studies mentioned above. In either case such differences mean that fouling may be difficult to correct for and, as such, should be taken into account for deployments where it is a significant issue, for example, if sampling rates used are from calibrations conducted in pure water and sampling takes place in wastewater. Additional experiments testing uptake in fouled and nonfouled samplers for a range of target compounds, preferably carried out in the environment, are required to determine the importance of fouling on POCIS uptake. Fouling has also been suggested as an issue for other configurations of PSDs suitable for sampling polar compounds, for example, when Empore disk–based receiving phases are exposed without membranes 10, 64.

Salinity, pH, and DOM

The effect of salinity on POCIS sampling rates will be highly compound-specific according to the various chemical groups that make up the compound. For example, Togola and Budzinski 43 found that acidic compounds showed no significant change in Rs with increasing salinity, whereas basic compounds including several tricyclic antidepressants showed reductions of up to 64%. Increasing salinity increases the energy required for a solute molecule cavity to form and increases the partitioning of neutral compounds toward nonaqueous phases, the so-called salting-out effect. This effect increases with the size and decreases with the polarity of the solute molecule 65 and, thus, should not be significant for most POCIS target compounds. Nevertheless, the application of freshwater-derived sampling rates to the marine environment should therefore be carried out with caution, despite suggestions to the contrary 48. Similarly, pH will affect sampling rates dependent on the species, with up to threefold increases for basic compounds with increasing alkalinity (pH 3–9) and decreases for acidic compounds, shown by Li et al. 39. Sampling rates for the neutral drug carbamazepine and the phenolic compounds studied (pKa values >10) were not affected 39. The same study also showed little advantage of replacing HLB with ion-exchange sorbents, in neutral waters, and no statistical difference in sampling rates with increases in DOM, albeit over a narrow range of DOM concentrations (3–5 mg/L). Furthermore, in a recent study, Charlestra et al. 58 also noted no apparent effect of natural organic matter concentrations (0.1–5 mg/L as total organic carbon) on POCIS Rs.

Properties of target compounds and sampler configurations

As well as the environmental conditions during sampler deployment listed above that affect uptake in POCIS, the physicochemical properties of individual compounds obviously will determine whether they accumulate or not and, if so, to what extent. In stark contrast to apolar passive samplers, where uptake has successfully been related to hydrophobicity 5 and molecular weight or volume 5, there are currently no such models to describe uptake in POCIS. This is due to the diversity of functional groups that polar compounds may possess and, therefore, the associated complexity of solute–solvent–sorbent interactions that are possible 50, 66.

Several attempts have been made to correlate observed sampling rates with some property of target compounds in the hope of being able to predict them for similar compounds, for example, using polynomial-type relationships similar to those observed for hydrophobic samplers to describe sampling rates for pesticides in terms of their log KOW 30, 35. However, other studies have found no such observable trend for alkylated phenols 46 or for a range of EDCs in either the pesticide or pharmaceutical version of POCIS 32. MacLeod et al. 38 found no overall trend for their studied target compounds based on log KOW, although improvements were apparent when compounds were separated based on charge speciation, where gaussian trends were fitted to anions and zwitterions and either a linear or curvilinear trend to cations depending on flow conditions. Gaussian trends were also fitted to a very limited number of sampling rates of beta-blockers and nonsteroidal anti-inflammatory drugs 38. Togola and Budzinski 43 reported a positive trend with log KOW for neutral and basic pharmaceuticals in freshwater but not for acidic compounds or in saline waters. Similarly, Li et al. 40 described a linear relationship with log KOW for a range of PPCPs and EDCs, when acidic and phenolic compounds were excluded. A linear relationship was also observed between Rs and chromatographic retention times on a C18 reversed-phase column in that study and subsequently when log KOW was normalized to the fraction of the neutral species, log D 39. Several of these results appear to be substantiated by recent work describing the predominance of apolar interactions for sorption of neutral compounds to Oasis HLB 50. Bartelt-Hunt et al. 39 hypothesized that if uptake is controlled by diffusion across the boundary layer, then it should be able to be modeled using the molecular weight of the compound, although no discernible trend was observed in the study. Due to lack of data, we have not discussed the relationship between sampling rates and target compound properties in terms of the two different types of POCIS available. Thus, it is important to note that such relationships may be specific to each sorbent, for example, where highly polar compounds have poor affinity for HLB but are retained by pesticide POCIS.

Affinity for the sorbent is not the only consideration, however; interactions with the PES membrane, which is normally discarded and not analyzed, are also apparent. These interactions have been reported several times for explaining a lag phase in uptake, which in general appears to be related to hydrophobicity 46, 60. This has been confirmed in more comprehensive studies where the membrane was also analyzed during calibration experiments 8. Thus, a three-compartment model (water–membrane–sorbent) might be sufficient to explain overall uptake in POCIS, similar to that proposed for SPMDs 8, 55. What is clear is that trying to extract general trends in uptake rates in relation to the physicochemical properties of target compounds based on the available data is not straightforward. This is not least due to differences between the calibration techniques used. For example, the apparent positive correlation between Rs and log KOW in some static depletion experiments mentioned above is likely due to overestimation of sampling rates for less polar compounds which have affinity for PES. In this regard, water–PES diffusion coefficients should also be determined as part of the calibration process (see also recommendations for calibrations above, Outlook for POCIS calibrations). A further unanswered issue is that of competitive sorption, whether compounds weakly retained by POCIS sorbents can be replaced by more strongly binding ones. In such situations, uptake could be under “matrix control” rather than WBL or membrane control. Thus, uptake experiments also need to be conducted with different concentrations of interfering substances present.

The surface area of a passive sampler is a crucial factor in the amount of analyte accumulated 16. As sampling rates for POCIS tend to be given as liters per day for a “standard” configuration, which has been defined as 41 cm2 3, deviations from this surface area will result in different sampling rates. Apart from the original descriptions of POCIS, which used a smaller sampler with a surface area of 18 cm2 2, 67, some studies report 42 cm2 68 and some 46 cm2 25, 69, which is a difference of up to 10% from the proposed standard configuration. Zhang et al. 48 found a positive relationship between POCIS exposure area (∼6–23 cm2) and Rs, although its effect appeared to be compound-specific (n = 4). The effects on uptake of the POCIS membrane surface area and the amount of sorbent used warrant further investigation for a range of target compounds.

Reported sampling rates

It is apparent, based on the above discussions, that there are many different experimental, compound-specific, and exposure-specific factors that affect the published sampling rates. Thus, sampling rates available in the literature should be treated as approximations and not applied as definitive numbers to other studies at different locations under different exposure conditions without adequate examination of those differences. For these reasons a comprehensive listing of all published Rs values is of little value to the current review (for reference purposes, they are provided in Supplemental Data, Table S1). Sampling rates reported in the literature for stagnant or near stagnant calibration conditions range from 0.001 L/d for malachite green 33 to 1.34 L/d for the antifungicide triclosan 39, whereas for turbulent conditions they range from 0.003 L/d for the pharmaceutical metabolite metroprolol acid 31 to 2.46 L/d for the antiulcerative drug omeprazole 38. The median values for all reported Rs values are 0.18 and 0.19 L/d (stagnant and turbulent exposures, respectively). At the lower end of the reported range are compounds that are not suited to be sampled by standard configurations of POCIS, with poor affinity for the sorbent. Rates of several liters per day may appear erroneous compared to most studies (Fig. 2); however, omeprazole was not the only compound which displayed high sampling rates in the study by MacLeod et al. 38, with both triclosan and the antidepressive drug fluoxetine having Rs values >1 L/d. The stability of these and similar compounds in the exposure system is of critical importance, especially when static depletion–type calibrations are used with sampling rates derived from reductions in water concentrations. Additionally, based on recent evidence of increasing affinity with the PES membrane with increasing log KOW 8, sampling rates for relatively hydrophobic compounds are likely to have been overestimated in several of these depletion-type calibrations. To address this issue, side-by-side calibrations using the different methods would need to be carried out with identical exposure parameters for the same compounds. As such studies are lacking, it is difficult to know to what extent differences in reported sampling rates are influenced by interactions with the membrane in depletion experiments. However, Figure 2 clearly shows a tendency of higher sampling rates in studies using static depletion. Triclosan sampling rates have also been shown to be relatively high in two other studies using the static depletion calibration method, for example, 0.753 to 1.929 L/d (quiescent and turbulent, respectively) 40 and 1.01 to 1.34 L/d under static conditions with a range of pH and DOM concentrations 38. Unfortunately, no other calibration method has been used to determine POCIS sampling rates for triclosan. However, for both omeprazole and fluoxetine sampling rates of 0.007 to 0.03 and 0.012 to 0.086 L/d (18 cm2 surface area) were reported by Alvarez et al. 2 from renewal-type calibrations compared to 2.47 and 0.223 to 1.37 L/d given by MacLeod et al. 38 from depletion-type calibrations. Similarly, differences in literature sampling rates for some alkylphenols have recently been shown to vary by a factor of between 10 and 100, which seems unlikely to be explained purely by exposure differences in calibration studies 25. Conversely, some pesticides, such as atrazine, which have been calibrated in several different studies 8, 30, 33, 35–37, show very similar sampling rates, 0.25 ± 0.03 L/d (mean ± standard deviation, n = 6). This illustrates the problem with POCIS sampling rates, where the variation apparent in published values (Fig. 2) is as likely to come from study artifacts or exposure differences as it is due to differing physicochemical properties of target analytes without being able to discern between the two aspects.

thumbnail image

Figure 2. Range of sampling rates reported in selected literature for hormones, industrial chemicals, pesticides, and pharmaceuticals. Whiskers show minimum and maximum values. No differentiation between types of polar organic chemical integrative samplers (POCIS) used or calibration conditions (temperature, etc.). Asterisk (*) indicates studies using the static depletion calibration method. Data from Alvarez et al. 2 are reported for POCIS with a surface area of 18 cm2.

Download figure to PowerPoint

CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Development of a PRC approach for POCIS

Although POCIS has already been shown to be a good tool for investigative monitoring and offers advantages over grab sampling, it is clear from the above discussion that the issues of uptake modeling and exposure correction are central to the development of POCIS as a quantitative technique. This second issue has been resolved for hydrophobic passive samplers by the use of PRCs 6. These substances, which must not interfere with analytical procedures, are spiked into samplers prior to deployment; and as their dissipation shows isotropic kinetics analogous to uptake, they are used to estimate Rs values of target compounds in situ. Thus, investigations have been undertaken to develop a PRC approach for POCIS, built around the observation that some compounds could be released again after spiking them into the sorbent 30. The release of compounds after initial uptake has been found in both laboratory and field calibration studies 31, 47, although the reasons remain unexplained. In a later study, Mazzella et al. 51 reported that a compound with relatively high fugacity (deuterium-labeled atrazine-desisopropyl) from the Oasis HLB sorbent showed promise as a PRC, narrowing the gap between POCIS water concentration estimates and those made by automatic sampling methods. Based on these results, Lissalde et al. 36 used the same compound to provide “quite acceptable” in situ sampling rates by adjusting those gained from laboratory calibrations using the environmental adjustment factor approach. However, no quantitative analysis of the applicability of these sampling rates was provided. Such an approach may be problematic if there is a discrepancy between the factors controlling the release of deuterium-labeled atrazine-desisopropyl and those controlling the uptake of target compounds. Conversely, the pesticides studied displayed fairly similar laboratory sampling rates, 0.195 ± 0.039 L/d (average ± standard deviation, n = 32), which may reduce such differences 36. Other compounds have subsequently been suggested as potential PRCs based on observational losses, although there were no attempts to apply them for sampling rate correction 25. Dias and Poole 66 demonstrated in an earlier mechanistic study of Oasis HLB that it showed both the properties of an adsorbent as well as a contribution of some partitioning mechanism, which may increase the likelihood of usable PRCs. It is interesting to note that in both of the available studies to examine the PRC approach in POCIS 30, 51 the Oasis containing pharmaceuticals POCIS version was used. Despite the triphasic pesticides POCIS configuration being considered the generic version, it has been shown to have a tendency to bind certain compounds very strongly due to the carbonaceous component, resulting in poor recoveries 3. Thus, it appears logical that the PRC approach may have limited applicability for the POCIS pesticides version. When using a different type of polar passive sampler based on Empore disks, Shaw et al. 71 attempted a PRC approach but found differences between the uptake stage of target compounds and the dissipation of the PRCs tested (linear vs nonlinear) and poor reproducibility.

Due to the range of conceivable interactions and based on the available calibration data, an all-encompassing PRC approach for POCIS may not be possible 72. Despite both the need for such a solution and some encouraging initial results, a more appropriate way forward is to first understand the processes governing uptake. Only then can an evaluation of the application of the PRC approach to POCIS be carried out adequately.

Surrogate PRC systems based around hydrophobic samplers

The codeployment of PRC-spiked hydrophobic samplers alongside POCIS has been suggested as a surrogate method for sensing sampling rate changes 3. Although there are many studies which codeploy samplers with different selectivity to measure a broad range of target compounds (see below, Reported Applications and Use), few have specifically corrected sampling rates of polar samplers with PRC data from hydrophobic ones. This is in part due to the need for cocalibration of both samplers, which is problematic for static-type experiments due to the high sampling rates of hydrophobic samplers. Harman et al. 46 calibrated POCIS and SPMDs together for alkylated phenols and polycyclic aromatic hydrocarbons, respectively, using a large flow-through system. Subsequently, both sampler types were exposed to measure the concentrations of those compounds in the receiving waters of discharges from the offshore oil industry 68, 73. Thus, the dissipation of PRCs in SPMDs during the calibration study could be compared to that during the environmental exposure. In this way, an environmental adjustment factor 70 was obtained, which was then used to correct POCIS Rs values from the calibration study. However, a suite of such calibration studies under a range of exposure conditions is required to validate this approach. One significant problem here is that dissipation of PRCs from nonpolar samplers is likely to be under WBL control, whereas the uptake of certain compounds in POCIS may be controlled by interactions with the membrane and/or the WBL 8. A further issue is likely to be physical differences between the currently available hydrophobic samplers and POCIS. Differences in the configuration of the same sampler (i.e., varying the depth of the sampler body) have been shown to affect Rs by as much as twofold for hydrophobic compounds 74. This approach may therefore be most easily realized when using the Chemcatcher passive sampling device as the deployment equipment and, thus, the exposure surface areas are the same for both hydrophobic and hydrophilic samplers 45, 75. Shaw et al. 76 used sampling rates from SPMDs and silicone rubber samplers to confirm similar sampling rates for pesticides between exposure stations for codeployed Chemcatcher samplers (SDB-RPS Empore disks covered with a PES membrane). However, no correction of the laboratory-derived sampling rates for environmental conditions was possible.

Novel approaches for POCIS performance reference systems

Another external correction method that has been proposed to correct for temperature and flow effects is based on studying the dissolution of a cast of calcium sulfate decahydrate, which is deployed alongside samplers. This approach has been applied to other passive sampling devices 77–79 and, thus, may offer some potential for use with POCIS, although this remains to be tested. Similar to using surrogate hydrophobic samplers, this method requires the plaster casts to be cocalibrated with the samplers, under a range of exposure conditions.

Instead of constructing a sampler with a rate-limiting membrane, a gel-based layer similar to that applied in diffusive gradient in thin film (DGT) samplers 80 widely used to measure inorganic compounds, may be used along with a suitable receiving phase 44. As in a DGT, the resistance to uptake is dominated by the gel layer so that uptake is much less affected by changes in water flow rates, reducing the need for exposure correction. Such an approach was recently reported as suitable for sampling the antibiotic sulfamethoxazole under a range of water pH values and ionic strengths 81. The sampling rate provided by this new organic-DGT was low (0.011 L/d) when compared to those reported for the same compound in POCIS (0.210–0.348 L/d) 40, 82, although this can likely be improved by increasing the surface area of the organic-DGT 81. The use of other materials for sampling polar compounds is possible, for example, silicone rubber, which has been shown to be able to sample several groups of medium polar compounds 83. However, the problem of sampling very hydrophilic compounds is likely to remain. Polyoxymethylene has also been claimed to be suitable for use as a passive sampler for many polar organic compounds 84. However, the usefulness of applying equilibrium passive sampling for the measurement of highly fluctuating concentrations of polar compounds, for example, in wastewater, appears limited. In summary, all these approaches may at least help toward uptake correction for certain applications. However, the overall inability to adjust uptake rates of polar passive samplers for differing environmental conditions during deployment means that estimations of time-integrated water concentrations of target compounds remain currently semiquantitative. If a custom solution is required for each study and suits only a limited group of similar compounds, then the overall applicability of polar samplers may be reduced. Thus, this issue requires considerable attention in future research if the full potential of the technique is to be realized. As mentioned previously, the first step of this process should be to focus on gaining an understanding of the processes governing uptake and how these are affected by exposure conditions.

REPORTED APPLICATIONS AND USE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Types of compounds

Over 300 individual compounds have been detected in POCIS (Fig. 2; Supplemental Data, Table S1), either in laboratory calibrations, in environmental exposures, or in both (note not all calibrated compounds have been detected in the environment). Of these, four of the most commonly measured compound groups are pesticides (>100 individual compounds in ≥26 studies), pharmaceuticals (>90 individual compounds in ≥21 studies), industrial chemicals (>30 individual compounds in ≥23 studies), and hormones (>15 individual compounds in ≥18 studies; see Supplemental Data, Table S1 for compound groups). Within these four groups, the most commonly measured compounds are the herbicide atrazine (20 studies), the stimulant caffeine (15 studies), the plastics production chemical bisphenol A (12 studies), and estrone (a metabolite of the hormone 17β-estradiol, 15 studies). Therefore, these four compounds appear well suited to be included in all future calibrations, which would facilitate the comparison of sampling rates between studies. The inclusion of compounds that are representative of specific chemical groups, for example, based on charge or interaction with the PES membrane, should also be considered.

Another significant compound group measured in POCIS is alkylated phenols (>60 individual compounds) both discharged in production waters in offshore oil and gas extraction 46, 47, 68, 73 and present in the environment from other sources 85. The latter also includes various nonyl and octylphenol ethoxylates and or their degradation products 32, 86–92, which are both included in the group industrial compounds above. Numerous compounds which are additives in a variety of PCPs have also been detected in POCIS extracts, for example, musks such as galaxolide and tonalide 85, 87, 93, 94, various UV filters 95, 96, and the insect repellent N,N-diethyl-meta-toluamide 82, 86–88, 93, 97, 98. Additionally, several hydrophobic compounds which are outside of the range of log KOW (<3–4) to which POCIS is normally applied 3 are reported to accumulate, such as polycyclic aromatic hydrocarbons 88, 99 and even polybrominated diphenylethers 99. If such compounds are targets to be measured, then much lower detection limits are likely to be reached by using hydrophobic PSDs with a larger surface area such as SPMDs, LDPE, or silicone rubber. Finally, several naturally occurring compounds have been measured in POCIS, notably cyanobacterial toxins microcystins 34, 100 and various natural fragrances or odors 87, 88.

Types of POCIS used

Consideration should be given before the start of a new study regarding which type of POCIS is to be used. For general screening, it is recommended to deploy both types to increase the range of compounds accumulated 3. The importance of carrying out some preliminary work considering POCIS sorbents and target compounds should not be underestimated. For example, at least half of studies concerning pesticide measurements are made using the pharmaceutical version of POCIS, for example, herbicides measured by Mazzella and coworkers 30, 51. Conversely, sampling rates for some pharmaceuticals have been shown to be higher in pesticide POCIS 59. Although those authors argue that there are practical reasons for choosing the pharmaceutical POCIS version based on the OASIS sorbent, not least where POCIS is constructed in house because the procedure for the pesticides POCIS version is more complicated (triphasic adsorbent mixture). Equally, consideration must be given to the suitability of the elution solvents used to extract POCIS sorbents. For example, Harman et al. 99 failed to measure mono- and dibutyltin when eluting Oasis with methanol, whereas recovery would have been improved by acidifying the solvent, facilitating calibration 101. A few studies report such preliminary work, for example, Arditsoglou and Voutsa 32, who improved recovery of EDCs from Oasis by using acetone instead of methanol. It is recommended that one should check the suitability of POCIS sorbents (and existing elution solvents) for target compounds before calibration/exposure. This can be carried out relatively easily using standard solid-phase extraction method development but is generally overlooked.

Other sorbents that have been tried in versions of POCIS include the polymeric reverse phase Strata X 102. Those authors claim this sorbent was applicable to a wider range of compounds than was Oasis HLB in preliminary trials, although they do not provide the data. Li and coworkers 39 calibrated POCIS containing the Oasis MAX and MCX (mixed mode anion exchange, and mixed mode cation exchange, respectively), as described previously. They concluded that the MAX and MCX versions showed little advantage over HLB in natural waters. Kaserzon et al. 103 used a weak anion exchange sorbent (Strata XAW) for sampling perfluorinated alkyl carboxylates and sulfonates, based on its superiority for sequestering the shorter-chain compounds. Other types of membrane-covered, sorbent-based samplers for polar compounds are also described in the literature, for example, coconut charcoal in teabag filter papers for nitrosamines 104 and sorbent-based samplers to target microalgal biotoxins 105–107.

Types of aquatic environments sampled

The majority of published POCIS studies making measurements in the environment are from a variety of lotic and lentic freshwater systems, with only a few studies considering purely marine sites 33, 68, 73, 108, 109. Interestingly, several studies report contamination in cave streams 88, 98 and even a coastal aquifer system 110. Dougherty et al. 97 also deployed POCIS in wells to consider pollution of groundwater. For studies carried out in the marine environment, consideration should be given where freshwater-derived sampling rates are used to estimate water concentrations in the marine environment as sampling rates may differ considerably between the two, as mentioned previously 43. In fact, sampling rates from laboratory conditions using tap or nanopure water have been used to estimate water concentrations in marine environments, without considering the effects of salinity 32. POCIS has also been applied to various measurement scenarios inside and around wastewater-treatment systems. Some examples are as follows: examination of the loadings, trends, comparisons, and fate of achiral and chiral pharmaceuticals 69; measuring steroidal estrogens and their removal 42; attempts to link chemical analysis with bioassay toxicity results from over 20 treated sewage effluents 111; calibration-type studies 31, 59; evaluating the presence of endocrine-disrupting substances posttreatment along with freshwater mussels 90; and studying the occurrence of illicit and therapeutic pharmaceuticals in wastewater effluent and surface waters 82.

Deployment times

In general, POCIS deployments tend to be for a period of several weeks, with the shortest reported deployments being 7 d 32, 82, 112. Such a time period is desirable to provide a reasonable time integrative window and secondly to aid detection at typical environmental concentrations. However, detection of illicit drugs in wastewater by POCIS was shown to be possible after just 3 d 31. Many environmental studies refer to the original POCIS work, where linear uptake was shown in the laboratory over a 56-d period at concentrations of 1 µg L/d, as evidence that field exposures where deployments are typically two to four weeks will also have experienced linear uptake. While this may appear logical, assuming environmental concentrations of typical target compounds are well below the 1 µg L/d level used for calibration, many other compounds (matrix) will accumulate during environmental exposures when compared to the laboratory. These additional compounds are not analyzed but may affect uptake in terms of competitive sorption processes. Care must thus be taken when assuming linearity of uptake during long deployments as this remains to be tested. For example, equilibrium or some kind of sorbent capacity may well have been reached for various estrogenic compounds during deployments of up to 169 d in Lake Thun in Switzerland 113. Fouling may also become an increasing problem during long deployments, although its effects on uptake remain poorly described for polar samplers. Where such long-term data are required, consecutive deployments of POCIS are likely to be more appropriate, although both the sampling and analysis burden are inherently increased. At least three such studies exist, two sampling wastewater continuously for 367 and 271 d 31, 69 and one year-long study of micropollutants from a Spanish fish farm 33, 108, allowing tends in contaminant concentrations to be elucidated.

Correlation with spot sampling

Many studies report comparisons between POCIS-generated water concentrations and those shown with spot sampling 36. Although this appears a logical step in the validation of (especially) sampling rates 2, it is difficult to know what information can be gained by comparing a three- to four-week POCIS deployment to a single grab sample. The accuracy of the water concentrations derived from POCIS will depend on the sampling rates applied (differences in exposure conditions, Factors Affecting POCIS Sampling Rates) and the ability of POCIS to integrate concentration changes adequately, as examined by Mazzella et al. 57 in controlled laboratory conditions. The verification of this time-integrated measurement by water analysis may require a sampling frequency beyond that which is practically feasible for certain scenarios such as wastewater-treatment systems 52. This potential variation in water concentrations which may be gained from spot sampling, one of the fundamental reasons for proposing passive sampling in the first place, means that for studies using a single or very few water samples the comparison would likely give different results if another sampling time point was used. Thus, comparisons will be poorest where the water-sampling interval is large and the water concentrations change rapidly 52, 57. As mentioned previously, detailed studies concerning the response time of POCIS to changes in water concentrations are lacking, and efforts should be focused on tackling this issue in a systematic way instead of ad hoc comparisons. This may be carried out in the laboratory or in situ where accumulations in samplers during overlapping exposures are statistically compared to frequent grab samples, under a range of fluctuating concentration scenarios.

Use of POCIS extracts in bioassays

Passive sampling devices, including POCIS, were initially developed for time-integrative monitoring of individual compounds measured by means of chemical analysis. However, PSD extracts have also been applied to biological testing, in both in vitro and in vivo assays 114. One of the first published POCIS studies 91 already covered the evaluation of POCIS extracts in a yeast-based assay to measure environmental estrogens. This study clearly established the potential of the use of POCIS in combination with bioassays. A large number of studies have now been performed where POCIS was combined with bioassays, many of them having dealt with endocrine disruptors, particularly the environmental estrogens 41, 90, 92, 102, 115–120. Other bioassays that have been used in combination with POCIS extracts have end points such as the inhibition of bacterial bioluminescence (i.e., an indicator of general toxicity 111, 115) and the inhibition of photosynthetic yield in algae or biofilms (i.e., an indicator of photosystem II inhibitors such as diuron and atrazine 111, 121).

Testing POCIS extracts in biological assays is useful for a number of reasons. First, POCIS continuously samples chemicals over the deployment period, and biologically active compounds that occur at varying concentrations can be more effectively monitored with POCIS than by means of grab sampling. Second, bioassays often require replication and dose–response curves for proper data evaluation 121. Therefore, a substantial original sample is needed to reach effective doses in the media used in bioassays. The fact that POCIS has sampling rates of one liter or more per week is thus an advantage, and POCIS can provide large enough samples for bioassays. In fact, samples from POCIS are often large enough to be used for more than one bioassay and multiple chemical analyses 111, 115. Third, it can be argued that the use of POCIS, rather than relying on traditional water sampling, is more relevant from an ecotoxicological perspective in that a passive sampler mimics uptake of compounds by organisms. For example, it was shown that the amounts of estrogenic activity in POCIS extracts correlated with the amounts of estrogenic activity found in the bile of fish 118. Therefore, using the time-integrative aspect of POCIS and combining POCIS extracts with multiple biological and chemical analyses provides for an efficient tool for investigative monitoring 92, 111.

One major strength of using bioassays to test environmental samples is that they can provide an integrative measure of the toxic potential of a group of compounds including unknown toxicants. This has been shown to be valid also when testing POCIS extracts in bioassays. Examples are (1) herbicides whose mode of action is to inhibit photosystem II 111 and (2) compounds that can bind to receptors in vertebrate cells 92. Unfortunately, the fact that a bioassay result comprises the toxicity of a range of compounds, each with its own toxic potency and its own sampling rate, thwarts the possibility of calculating an accurate time-weighted average (TWA) concentration for the measured toxicity in a PSD. For example, a toxic potential of 100 ng per POCIS may equate to 1 L of sampled water containing a potent chemical (i.e., a TWA of 100 ng/L) or it may equate to 20 L of sampled water containing a 10-fold less potent chemical with a twofold lower sampling rate (a TWA of 5 ng/L). Thus, it is not possible to calibrate POCIS to provide TWA concentrations for toxicity assessments. The issue also highlights a major critical question that emerges when combining a POCIS extract with a bioassay: How well does the mixture of chemicals in the extract resemble the mixture of chemicals present in the water during deployment? For the purpose of quantifying individual compounds, it is not a problem when each compound has a different sampling rate. As long as the sampling rate of a compound is known, a TWA concentration may be calculated. However, when a number of compounds with the same mode of action have different sampling rates this causes problems. A compound may dominate the toxicity in a water body, but because of its low sampling rate, it will be underrepresented in a POCIS extract to the extent where it may no longer dominate toxicity. Given the observed variability in sampling rates (Fig. 2), it is thus inevitable that the mixture of toxicants in the POCIS extract will differ from that in the water where the POCIS was deployed and, thus, is not a fully representative sample. Although this issue also applies to an enrichment technique such as solid-phase extraction, it is easier to optimize, standardize, and control the enrichment step with solid-phase extraction, at least for known toxicants.

This critical issue of shifting toxicity profiles between the sampled water and POCIS extracts has not yet been examined, to our knowledge, but can be illustrated for a well-studied class of compounds like the environmental estrogens. Zhang et al. 48 found that POCIS sampled four steroidal estrogens in a narrow range of 0.04 to 0.05 L/d (for a 27-mm POCIS). This suggests that the estrogenic toxicity profile would be preserved when going from the sampled water to the POCIS extract (see also Rujiralai et al. 42). However, Arditsoglou and Voutsa 32 looked at a broader range of endocrine disruptors and found that the potent steroidal EE2 had an almost twofold higher sampling rate than five other steroidal estrogens. Furthermore, the endocrine disruptor 4-octylphenol had a 30-fold lower sampling rate than EE2. Going by these sampling rate data 32, the toxic potential of a POCIS extract would contain an overrepresentation of the pharmaceutical EE2 and would largely miss the contribution of the industrial chemical 4-octylphenol. Currently, there is no way around this problem. Consequently, when combining POCIS with biological analysis, it is advisable to be aware of the issue and to check if sampling rates of know toxicants are in a similar range.

CONCLUSIONS AND RECOMMENDATIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Polar organic chemical integrative sampler has been applied to a wide variety of over 300 target compounds, different environments, and sampling scenarios. In several of these studies it is difficult to see how the samples could have been taken otherwise. However, the results have been examined at different levels of complexity. Despite the advantages over traditional spot sampling offered by POCIS, the two interrelated issues of modeling uptake and exposure correction remain unresolved. These questions will require considerable attention before we can approach quantitative measurements for polar passive samplers. Although POCIS already serves as a versatile, economic, and robust tool for investigative monitoring studies and for observing spatial and temporal trends, sampling rates are not yet robust enough to supply reliable TWA concentrations, particularly for the monitoring of environmental quality standards. We make the following recommendations based on our review of the literature. First, preliminary studies considering sorbents and extraction methods/solvents for the target compounds should be carried out before sampler calibration and/or environmental exposure; this should include examining partitioning to PES. In addition, laboratory calibrations need standardizing, including methods for measuring and reporting as well as the calculation of sampling rates and the frequency of sampling for water concentrations. Furthermore, though static calibrations are very economical, flow-through calibrations are methodologically better approximations of POCIS deployments in the field. Static depletion calibration should be used only with adequate consideration of uptake by the PES membrane, and where possible, widely studied compounds should be added to all calibration experiments (e.g., atrazine). Next, more mechanistic-type studies are required to understand the adsorption process and transport through/into the membranes and which factors are controlling uptake. This should also include examination of competitive sorption by interfering substances: sampling rates should be seen as a guide, rather than a definite number. Finally, providing a TWA for toxicity measurements made in POCIS extracts should not be attempted as toxicity involves mixtures of compounds with different toxic potencies and different sampling rates. When interpreting bioassay results from POCIS extracts it has to be kept in mind that POCIS will not provide a one-to-one mirror image of the cocktail of toxicants in the sampled water.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information
  • 1
    European Commission. 2012. Proposal for a directive of the European Parliament and of the Council amending Directives 2000/60/EC and 2008/105/EC as regards priority substances in the field of water policy. Brussels, Belgium.
  • 2
    Alvarez DA, Petty JD, Huckins JN, Jones-Lepp TL, Getting DT, Goddard JP, Manahan SE. 2004. Development of a passive, in situ, integrative sampler for hydrophilic organic contaminants in aquatic environments. Environ Tocicol Chem 23: 16401648.
  • 3
    Alvarez DA, Huckins JN, Petty JD, Jones-Lepp TL, Stuer-Lauridsen F, Getting DT, Goddard JP, Gravell A. 2007. Tool for monitoring hydrophillic contaminants in water: Polar organic chemical integrative sampler (POCIS). In Greenwood R, Mills G, Vrana B, eds, Passive Sampling Techniques in Environmental Monitoring, Vol 48—Comprehensive Analytical Chemistry. Elsevier, Amsterdam, The Netherlands, pp 171197.
  • 4
    Huckins JN, Petty JD, Booij K. 2006. Monitors of Organic Chemicals in the Environment. Springer, New York, NY, USA.
  • 5
    Booij K, Vrana B, Huckins JN. 2007. Theory, modeling, and calibration of passive samplers used in water monitoring. In Greenwood R, Mills G, Vrana B, eds, Passive Sampling Techniques in Environmental Monitoring, Vol 48—Comprehensive Analytical Chemistry. Elsevier, Amsterdam, The Netherlands, pp 141169.
  • 6
    Booij K, Sleiderink HM, Smedes F. 1998. Calibrating the uptake kinetics of semipermeable membrane devices using exposure standards. Environ Toxicol Chem 17: 12361245.
  • 7
    Huckins JN, Petty JD, Orazio CE, Lebo JA, Clark RC, Gibson VL, Gala WR, Echols KR. 1999. Determination of uptake kinetics (sampling rates) by lipid-containing semipermeable membrane devices (SPMDs) for polycyclic aromatic hydrocarbons (PAHs) in water. Environ Sci Technol 33: 39183923.
  • 8
    Vermeirssen ELM, Dietschweiler C, Escher BI, van der Voet J, Hollender J. 2012. Transfer kinetics of polar organic compounds over polyethersulphone membranes in the passive samplers POCIS and Chemcatcher. Environ Sci Technol 46: 67596766.
  • 9
    Vermeirssen ELM, Asmin J, Escher BI, Kwon JH, Steimen I, Hollender J. 2008. The role of hydrodynamics, matrix and sampling duration in passive sampling of polar compounds with Empore SDB-RPS disks. J Environ Monit 10: 119128.
  • 10
    Schafer RB, Paschke A, Liess M. 2008. Aquatic passive sampling of a short-term thiacloprid pulse with the Chemcatcher: Impact of biofouling and use of a diffusion-limiting membrane on the sampling rate. J Chromatogr A 1203: 16.
  • 11
    O'Brien D, Bartkow M, Mueller JF. 2011. Determination of deployment specific chemical uptake rates for SDB-RPD Empore disk using a passive flow monitor (PFM). Chemosphere 83: 12901295.
  • 12
    Stephens BS, Kapernick AP, Eaglesham G, Mueller JF. 2009. Event monitoring of herbicides with naked and membrane-covered Empore disk integrative passive sampling devices. Mar Pollut Bull 58: 11161122.
  • 13
    Seethapathy S, Gorecki T, Li XJ. 2008. Passive sampling in environmental analysis. J Chromatogr A 1184: 234253.
  • 14
    Kot-Wasik A, Zabiegala B, Urbanowicz M, Dominiak E, Wasik A, Namiesnik J. 2007. Advances in passive sampling in environmental studies. Anal Chim Acta 602: 141163.
  • 15
    Zabiegala B, Kot-Wasik A, Urbanowicz M, Namiesnik J. 2010. Passive sampling as a tool for obtaining reliable analytical information in environmental quality monitoring. Anal Bioanal Chem 396: 273296.
  • 16
    Vrana B, Mills GA, Allan IJ, Dominiak E, Svensson K, Knutsson J, Morrison G, Greenwood R. 2005. Passive sampling techniques for monitoring pollutants in water. Trac Trend Anal Chem 24: 845868.
  • 17
    Stuer-Lauridsen F. 2005. Review of passive accumulation devices for monitoring organic micropollutants in the aquatic environment. Environ Pollut 136: 503524.
  • 18
    Esteve-Turrillas FA, Pastor A, Yusa V, de la Guardia M. 2007. Using semi-permeable membrane devices as passive samplers. Trend Anal Chem 26: 703712.
  • 19
    Esteve-Turrillas FA, Yusa V, Pastor A, de la Guardia M. 2008. New perspectives in the use of semipermeable membrane devices as passive samplers. Talanta 74: 443457.
  • 20
    Mills GA, Vrana B, Allan I, Alvarez DA, Huckins JN, Greenwood R. 2007. Trends in monitoring pharmaceuticals and personal-care products in the aquatic environment by use of passive sampling devices. Anal Bioanal Chem 387: 11531157.
  • 21
    Wong CS, MacLeod SL. 2009. JEM Spotlight: Recent advances in analysis of pharmaceuticals in the aquatic environment. J Environ Monit 11: 923936.
  • 22
    Buchberger WW. 2011. Current approaches to trace analysis of pharmaceuticals and personal care products in the environment. J Chromatogr A 1218: 603318.
  • 23
    Soderstrom H, Lindberg RH, Fick J. 2009. Strategies for monitoring the emerging polar organic contaminants in water with emphasis on integrative passive sampling. J Chromatogr A 1216: 623630.
  • 24
    Petrovic M, Farre M, de Alda ML, Perez S, Postigo C, Kock M, Radjenovic J, Gros M, Barcelo D. 2010. Recent trends in the liquid chromatography-mass spectrometry analysis of organic contaminants in environmental samples. J Chromatogr A 1217: 40044017.
  • 25
    Miege C, Budzinski H, Jacquet R, Soulier C, Pelte T, Coquery M. 2012. Polar organic chemical integrative sampler (POCIS): Application for monitoring organic micropollutants in wastewater effluent and surface water. J Environ Monitor 14: 626635.
  • 26
    Yates K, Davies I, Webster L, Pollard P, Lawton L, Moffat C. 2007. Passive sampling: Partition coefficients for a silicone rubber reference phase. J Environ Monit 9: 11161121.
  • 27
    Smedes F, Geertsma RW, van der Zande T, Booij K. 2009. Polymer–water partition coefficients of hydrophobic compounds for passive sampling: Application of cosolvent models for validation. Environ Sci Technol 43: 70477054.
  • 28
    Rusina TP, Smedes F, Klanova J, Booij K, Holoubek I. 2007. Polymer selection for passive sampling: A comparison of critical properties. Chemosphere 68: 13441351.
  • 29
    Rusina TP, Smedes F, Koblizkova M, Klanova J. 2010. Calibration of silicone rubber passive samplers: Experimental and modeled relations between sampling rate and compound properties. Environ Sci Technol 44: 362367.
  • 30
    Mazzella N, Dubernet JF, Delmas F. 2007. Determination of kinetic and equilibrium regimes in the operation of polar organic chemical integrative samplers—Application to the passive sampling of the polar herbicides in aquatic environments. J Chromatogr A 1154: 4251.
  • 31
    Harman C, Reid M, Thomas KV. 2011. In situ calibration of a passive sampling device for selected illicit drugs and their metabolites in wastewater, and subsequent year-long assessment of community drug usage. Environ Sci Technol 45: 56765682.
  • 32
    Arditsoglou A, Voutsa D. 2008. Passive sampling of selected endocrine disrupting compounds using polar organic chemical integrative samplers. Environ Pollut 156: 316324.
  • 33
    Martinez Bueno MJ, Hernando MD, Aguera A, Fernandez-Alba AR. 2009. Application of passive sampling devices for screening of micro-pollutants in marine aquaculture using LC-MS/MS. Talanta 77: 15181527.
  • 34
    Kohoutek J, Marsalek B, Blaha L. 2010. Evaluation of the novel passive sampler for cyanobacterial toxins microcystins under various conditions including field sampling. Anal Bioanal Chem 397: 823828.
  • 35
    Thomatou A-A, Zacharias I, Hela D, Konstantinou I. 2011. Passive sampling of selected pesticides in aquatic environment using polar organic chemical integrative samplers. Environ Sci Pollut Res 18: 12221233.
  • 36
    Lissalde S, Mazzella N, Fauvelle V, Delmas F, Mazellier P, Legube B. 2011. Liquid chromatography coupled with tandem mass spectrometry method for thirty-three pesticides in natural water and comparison of performance between classical solid phase extraction and passive sampling approaches. J Chromatogr A 1218: 14921502.
  • 37
    Bartelt-Hunt SL, Snow DD, Damon-Powell T, Brown DL, Prasai G, Schwarz M, Kolok AS. 2011. Quantitative evaluation of laboratory uptake rates for pesticides, pharmaceuticals, and steroid hormones using POCIS. Environ Toxicol Chem 30: 14121420.
  • 38
    MacLeod SL, McClure EL, Wong CS. 2007. Laboratory calibration and field deployment of the polar organic chemical integrative sampler for pharmaceuticals and personal care products in wastewater and surface water. Environ Toxicol Chem 26: 25172529.
  • 39
    Li H, Helm PA, Paterson G, Metcalfe CD. 2011. The effects of dissolved organic matter and pH on sampling rates for polar organic chemical integrative samplers (POCIS). Chemosphere 83: 371280.
  • 40
    Li HX, Helm PA, Metcalfe CD. 2010. Sampling in the Great Lakes for pharmaceuticals, personal care products, and endocrine-disrupting substances using the passive polar organic chemical integrative sampler. Environ Toxicol Chem 29: 751762.
  • 41
    Matthiessen P, Arnold D, Johnson AC, Pepper TJ, Pottinger TG, Pulman KGT. 2006. Contamination of headwater streams in the United Kingdom by oestrogenic hormones from livestock farms. Sci Tot Environ 367: 616630.
  • 42
    Rujiralai T, Bull ID, Llewellyn N, Evershed RP. 2011. In situ polar organic chemical integrative sampling (POCIS) of steroidal estrogens in sewage treatment works discharge and river water. J Environ Monit 13: 14271434.
  • 43
    Togola A, Budzinski H. 2007. Development of polar organic integrative samplers for analysis of pharmaceuticals in aquatic systems. Anal Chem 79: 67346741.
  • 44
    Alvarez DA. 1999. Development of an integrative sampling device for hydrophillic organic contaminants in aquatic environments. PhD thesis. University of Missouri, Columbia, MO, USA.
  • 45
    Vrana B, Mills GA, Dominiak E, Greenwood R. 2006. Calibration of the Chemcatcher passive sampler for the monitoring of priority organic pollutants in water. Environ Pollut 142: 333343.
  • 46
    Harman C, Tollefsen KE, Bøyum O, Thomas KV, Grung M. 2008. Uptake rates of alkylphenols, PAHs and carbazoles in semipermeable membrane devices (SPMDs) and polar organic chemical integrative samplers (POCIS). Chemosphere 72: 15101516.
  • 47
    Harman C, Boyum O, Thomas KV, Grung M. 2009. Small but different effect of fouling on the uptake rates of semipermeable membrane devices and polar organic chemical integrative samplers. Environ Toxicol Chem 28: 23242332.
  • 48
    Zhang ZL, Hibberd A, Zhou JL. 2008. Analysis of emerging contaminants in sewage effluent and river water: Comparison between spot and passive sampling. Anal Chim Acta 607: 3744.
  • 49
    Meadows JC, Echols KR, Huckins JN, Borsuk FA, Carline RF, Tillitt DE. 1998. Estimation of uptake rate constants for PCB congeners accumulated by semipermeable membrane devices and brown trout (Salmo trutta). Environ Sci Technol 32: 18471852.
  • 50
    Bauerlein PS, Mansell JE, ter Lakk TL, de Voogt P. 2012. Sorption behavior of charged and neutral polar organic compounds on solid phase extraction materials: Which functional group governs sorption? Environ Sci Technol 46: 954961.
  • 51
    Mazzella N, Lissalde S, Moreira S, Delmas F, Mazellier P, Huckins JN. 2010. Evaluation of the use of performance reference compounds in an Oasis-HLB adsorbent based passive sampler for improving water concentration estimates of polar herbicides in freshwater. Environ Sci Technol 44: 17131719.
  • 52
    Ort C, Lawrence MG, Rieckermann J, Joss A. 2010. Sampling for pharmaceuticals and personal care products (PPCPs) and illicit drugs in wastewater systems: Are your conclusions valid? A critical review. Environ Sci Technol 44: 60246035.
  • 53
    Hawker DW. 2010. Modeling the response of passive samplers to varying ambient fluid concentrations of organic contaminants. Environ Toxicol Chem 29: 591596.
  • 54
    Gourlay-France C, Lorgeoux C, Tusseau-Vuillemin MH. 2008. Polycyclic aromatic hydrocarbon sampling in wastewaters using semipermeable membrane devices: Accuracy of time-weighted average concentration estimations of truly dissolved compounds. Chemosphere 73: 11941200.
  • 55
    Gale RW. 1998. Three-compartment model for contaminant accumulation by semipermeable membrane devices. Environ Sci Technol 32: 22922300.
  • 56
    Shaw M, Mueller JF. 2009. Time integrative passive sampling: How well do Chemcatchers integrate fluctuating pollutant concentrations? Environ Sci Technol 43: 14431448.
  • 57
    Mazzella N, Debenest T, Delmas F. 2008. Comparison between the polar organic chemical integrative sampler and the solid-phase extraction for estimating herbicide time-weighted average concentrations during a microcosm experiment. Chemosphere 73: 545550.
  • 58
    Charlestra L, Amirbahman A, Courtemanch DL, Alvarez DA, Patterson H. 2012. Estimating pesticide sampling rates by the polar organic chemical integrative sampler (POCIS) in the presence of natural organic matter and varying hydrodynamic conditions. Environ Pollut 169: 98104.
  • 59
    Li H, Vermeirssen ELM, Helm PA, Metcalfe C. 2010. Controlled field evaluation of water flow rate effects on sampling polar organic compounds using polar organic chemcial integrative samplers. Environ Toxicol Chem 29: 24612469.
  • 60
    Vermeirssen ELM, Bramaz N, Hollender J, Singer H, Escher BI. 2009. Passive sampling combined with ecotoxicological and chemical analysis of pharmaceuticals and biocides—evaluation of three Chemcatcher (TM) configurations. Water Res 43: 903914.
  • 61
    Vrana B, Schuurmann G. 2002. Calibrating the uptake kinetics of semipermeable membrane devices in water: Impact of hydrodynamics. Environ Sci Technol 36: 290296.
  • 62
    Booij K, Van Bommel R, Mets A, Dekker R. 2006. Little effect of excessive biofouling on the uptake of organic contaminants by semipermeable membrane devices. Chemosphere 65: 24852492.
  • 63
    Richardson BJ, Abbott S, McClellan KE, Zheng GJ, Lam PKS. 2008. The use of permeability reference compounds in biofouled semi-permeable membrane devices (SPMDs): A laboratory-based investigation. Mar Pollut Bull 56: 16631667.
  • 64
    Tan BLL, Hawker DW, Muller JF, Leusch FDL, Tremblay LA, Chapman HF. 2007. Comprehensive study of endocrine disrupting compounds using grab and passive sampling at selected wastewater treatment plants in south east Queensland, Australia. Environ Int 33: 654669.
  • 65
    Endo S, Pfennigsdorff A, Goss KU. 2012. Salting-out effect in aqueous NaCl solutions: Trends with size and polarity of solute molecules. Environ Sci Technol 46: 14961503.
  • 66
    Dias NC, Poole CF. 2002. Mechanistic study of the sorption properties of Oasis HLB and its use in solid-phase extraction. Chromatographia 56: 269275.
  • 67
    Jones-Lepp TL, Alvarez DA, Petty JD, Huckins JN. 2004. Polar organic chemical integrative sampling and liquid chromatography-electrospray/ion-trap mass spectrometry for assessing selected prescription and illicit drugs in treated sewage effluents. Arch Environ Contam Toxicol 47: 427439.
  • 68
    Harman C, Brooks S, Sundt RC, Meier S, Grung M. 2011. Field comparison of passive sampling and biological approaches for measuring exposure of PAH and alkylphenols from offshore produced water discharges. Mar Pollut Bull 63: 141148.
  • 69
    MacLeod SL, Wong CS. 2010. Loadings, trends, comparisons, and fate of achiral and chiral pharmaceuticals in wastewaters from urban tertiary and rural aerated lagoon treatments. Water Res 44: 533544.
  • 70
    Huckins JN, Petty JD, Lebo JA, Almeida FV, Booij K, Alvarez DA, Clark RC, Mogensen BB. 2002. Development of the permeability/performance reference compound approach for in situ calibration of semipermeable membrane devices. Environ Sci Technol 36: 8591.
  • 71
    Shaw M, Eaglesham G, Mueller JF. 2009. Uptake and release of polar compounds in SDB-RPS Empore disks: Implications for their use as passive samplers. Chemosphere 75: 17.
  • 72
    Harman C, Allan IJ, Bauerlein PS. 2011. The challenge of exposure correction for polar passive samplers—The PRC and the POCIS. Environ Sci Technol 45: 91209121.
  • 73
    Harman C, Farmen E, Tollefsen KE. 2010. Monitoring North Sea oil production discharges using passive sampling devices coupled with in vitro bioassay techniques. J Environ Monit 12: 16991708.
  • 74
    Lobpreis T, Vrana B, Dominiak E, Dercova K, Mills G, Greenwood R. 2007. Effect of housing geometry on the perfermance of Chemcatcher passive sampler for the monitoring of hydrophobic organic pollutants in water. Environ Pollut 153: 706710.
  • 75
    Tran ATK, Hyne RV, Doble P. 2007. Calibration of a passive sampling device for time-integrated sampling of hydrophilic herbicides in aquatic environments. Environ Toxicol Chem 26: 435443.
  • 76
    Shaw M, Furnas MJ, Fabricius K, Haynes D, Carter S, Eaglesham G, Mueller JF. 2010. Monitoring pesticides in the Great Barrier Reef. Mar Pollut Bull 60: 113122.
  • 77
    O'Brien D, Hawker D, Shaw M, Mueller JF. 2011. The performance of passive flow monitors and phosphate accumulating passive samplers when exposed to pulses in external water flow rate and/or external phosphate concentrations. Environ Pollut 159: 14351441.
  • 78
    O'Brien DS, Chiswell B, Mueller JF. 2009. A novel method for the in situ calibration of flow effects on a phosphate passive sampler. J Environ Monit 11: 212219.
  • 79
    Kira S, Nogami Y, Ito T, Hayatsu H. 1998. Measurement of a time-weighted average concentration of polycyclic aromatic hydrocarbons in seawater: An improved procedure of blue rayon hanging technique for monitoring benzo(a)pyrene. Mar Environ Res 46: 267271.
  • 80
    Zhang H, Davison W. 1995. Performance characteristics of diffusion gradients in thin films for the in-situ measurement of trace metals in aqueous solution. Anal Chem 67: 33913400.
  • 81
    Chen C-E, Zhang H, Jones KC. 2012. A novel passive water sampler for in situ sampling of antibiotics. J Environ Monit 14: 15231530.
  • 82
    Bartelt-Hunt SL, Snow DD, Damon T, Shockley J, Hoagland K. 2009. The occurrence of illicit and therapeutic pharmaceuticals in wastewater effluent and surface waters in Nebraska. Environ Pollut 157: 786791.
  • 83
    Wille K, Claessens M, Rappe K, Monteyne E, Janssen CR, De Brabander HF, Vanhaecke L. 2011. Rapid quantification of pharmaceuticals and pesticides in passive samplers using ultra high performance liquid chromatography coupled to high resolution mass spectrometry. J Chromatogr A 1218: 91629173.
  • 84
    Endo S, Hale SE, Goss K, Arp HPH. 2011. Equilibrium partition coefficients of diverse polar and nonpolar organic compounds to polyoxymethylene (POM) passive sampling devices. Environ Sci Technol 45: 1012410132.
  • 85
    Rosen MR, Alvarez DA, Goodbred SL, Leiker TJ, Patino R. 2010. Sources and distribution of organic compounds using passive samplers in Lake Mead national recreation area, Nevada and Arizona, and their implications for potential effects on aquatic biota. J Environ Qual 39: 11611172.
  • 86
    Writer JH, Barber LB, Brown GK, Taylor HE, Kiesling RL, Ferrey ML, Jahns ND, Bartell SE, Schoenfuss HL. 2010. Anthropogenic tracers, endocrine disrupting chemicals, and endocrine disruption in Minnesota lakes. Sci Total Environ 409: 100111.
  • 87
    Alvarez DA, Stackelberg PE, Petty JD, Huckins JN, Furlong ET, Zaugg SD, Meyer MT. 2005. Comparison of a novel passive sampler to standard water-column sampling for organic contaminants associated with wastewater effluents entering a New Jersey stream. Chemosphere 61: 610622.
  • 88
    Bidwell JR, Becker C, Hensley S, Stark R, Meyer MT. 2010. Occurrence of organic wastewater and other contaminants in cave streams in northeastern Oklahoma and northwestern Arkansas. Arch Environ Contam Toxicol 58: 286298.
  • 89
    Di Carro M, Scapolla C, Liscio C, Magi E. 2010. Development of a fast liquid chromatography-tandem mass spectrometry method for the determination of endocrine-disrupting compounds in waters. Anal Bioanal Chem 398: 10251034.
  • 90
    Liscio C, Magi E, Di Carro M, Suter MJF, Vermeirssen ELM. 2009. Combining passive samplers and biomonitors to evaluate endocrine disrupting compounds in a wastewater treatment plant by LC/MS/MS and bioassay analyses. Environ Pollut 157: 27162721.
  • 91
    Petty JD, Huckins JN, Alvarez DA, Brumbaugh WG, Cranor WL, Gale RW, Rastall AC, Jones-Lepp TL, Leiker TJ, Rostad CE, Furlong ET. 2004. A holistic passive integrative sampling approach for assessing the presence and potential impacts of waterborne environmental contaminants. Chemosphere 54: 695705.
  • 92
    Creusot N, Kinani S, Balaguer P, Tapie N, LeMenach K, Maillot-Marechal E, Porcher JM, Budzinski H, Ait-Aissa S. 2010. Evaluation of an hPXR reporter gene assay for the detection of aquatic emerging pollutants: Screening of chemicals and application to water samples. Anal Bioanal Chem 396: 569583.
  • 93
    Alvarez DA, Cranor WL, Perkins SD, Schroeder VL, Iwanowicz LR, Clark RC, Guy CP, Pinkney AE, Blazer VS, Mullican JE. 2009. Reproductive health of bass in the Potomac, USA, drainage: Part 2. Seasonal occurrence of persistent and emerging organic contaminants. Environ Toxicol Chem 28: 10841095.
  • 94
    Iwanowicz LR, Blazer VS, Guy CP, Pinkney AE, Mullican JE, Alvarez DA. 2009. Reproductive health of bass in the Potomac, USA, drainage: Part 1. Exploring the effects of proximity to wastewater treatment plant discharge. Environ Toxicol Chem 28: 10721083.
  • 95
    Fent K, Zenker A, Rapp M. 2010. Widespread occurrence of estrogenic UV-filters in aquatic ecosystems in Switzerland. Environ Pollut 158: 18171824.
  • 96
    Zenker A, Schmutz H, Fent K. 2008. Simultaneous trace determination of nine organic UV-absorbing compounds (UV filters) in environmental samples. J Chromatogr A 1202: 6474.
  • 97
    Dougherty JA, Swarzenski PW, Dinicola RS, Reinhard M. 2010. Occurrence of herbicides and pharmaceutical and personal care products in surface water and groundwater around Liberty Bay, Puget Sound, Washington. J Environ Qual 39: 11731180.
  • 98
    Fox JT, Adams G, Sharum M, Steelman KL. 2010. Passive sampling of bioavailable organic chemicals in Perry County, Missouri cave streams. Environ Sci Technol 44: 88358841.
  • 99
    Harman C, Bøyum O, Tollefsen KE, Thomas KV, Grung M. 2008. Uptake of some selected aquatic pollutants in semipermeable membrane devices (SPMDs) and the polar organic chemical integrative sampler (POCIS). J Environ Monit 10: 239247.
  • 100
    Kohoutek J, Babica P, Blaha L, Marsalek B. 2008. A novel approach for monitoring of cyanobacterial toxins: Development and evaluation of the passive sampler for microcystins. Anal Bioanal Chem 390: 11671172.
  • 101
    Gonzalez-Toledo E, Ortuno A, Compano R, Granados M, Prat MD. 2002. Solid-phase extraction-liquid chromatography-fluorimetry for organotin speciation in natural waters. Chromatographia 55: 1924.
  • 102
    Balaam JL, Grover D, Johnson AC, Jurgens M, Readman J, Smith AJ, White S, Williams R, Zhou JL. 2010. The use of modeling to predict levels of estrogens in a river catchment: How does modeled data compare with chemical analysis and in vitro yeast assay results? Sci Tot Environ 408: 48264832.
  • 103
    Kaserzon SL, Kennedy K, Hawker DW, Thompson J, Carter S, Roach AC, Booij K, Mueller JF. 2012. Development and calibration of a passive sampler for perfluorinated alkyl carboxylates and sulfonates in water. Environ Sci Technol 46: 49854993.
  • 104
    Kaserzon SL, Kennedy K, Hawker DW, Holling N, Escher BI, Booij K, Mueller JF. 2011. Development and calibration of a passive sampler for N-nitrosodimethylamine (NDMA) in water. Chemosphere 84: 497503.
  • 105
    MacKenzie LA. 2010. In situ passive solid-phase adsorption of micro-algal biotoxins as a monitoring tool. Curr Opin Biotechnol 21: 326331.
  • 106
    Rundberget T, Gustad E, Samdal IA, Sandvik M, Miles CO. 2009. A convenient and cost-effective method for monitoring marine algal toxins with passive samplers. Toxicon 53: 543550.
  • 107
    MacKenzie L, Beuzenberg V, Holland P, McNabb P, Selwood A. 2004. Solid phase adsorption toxin tracking (SPATT): A new monitoring tool that simulates the biotoxin contamination of filter feeding bivalves. Toxicon 44: 901918.
  • 108
    Munoz I, Bueno MJM, Aguera A, Fernandez-Alba AR. 2010. Environmental and human health risk assessment of organic micro-pollutants occurring in a Spanish marine fish farm. Environ Pollut 158: 18091816.
  • 109
    Harman C, Thomas KV, Tollefsen KE, Meier S, Boyum O, Grung M. 2009. Monitoring the freely dissolved concentrations of polycyclic aromatic hydrocarbons (PAH) and alkylphenols (AP) around a Norwegian oil platform by holistic passive sampling. Mar Pollut Bull 58: 16711679.
  • 110
    Metcalfe CD, Beddows PA, Bouchot GG, Metcalfe T, Li H, Van Lavieren H. 2011. Contaminants in the coastal karst aquifer system along the Caribbean coast of the Yucatan Peninsula, Mexico. Environ Pollut 159: 991997.
  • 111
    Vermeirssen ELM, Hollender J, Bramaz N, Van der Voet J, Escher BI. 2010. Linking toxicty in algal and bacterial assays with chemical analysis in passive samplers deployed in 21 treated sewage effluents. Environ Toxicol Chem 29: 25752582.
  • 112
    Sellin MK, Snow DD, Schwarz M, Carter BJ, Kolok AS. 2009. Agrichemicals in Nebraska, USA, watersheds: Occurrence and endocrine effects. Environ Toxicol Chem 28: 24432448.
  • 113
    Liedtke A, Schonenberger R, Eggen RIL, Suter MJF. 2009. Internal exposure of whitefish (Coregonus lavaretus) to estrogens. Aquat Toxicol 93: 158165.
  • 114
    Shaw M, Negri A, Fabricius K, Mueller JF. 2009. Predicting water toxicity: Pairing passive sampling with bioassays on the Great Barrier Reef. Aquat Toxicol 95: 108116.
  • 115
    Alvarez DA, Cranor WL, Perkins SD, Clark RC, Smith SB. 2008. Chemical and toxicologic assessment of organic contaminants in surface water using passive samplers. J Environ Qual 37: 10241033.
  • 116
    Burki R, Vermeirssen ELM, Korner O, Joris C, Burkhardt-Holm P, Segner H. 2006. Assessment of estrogenic exposure in brown trout (Salmo trutta) in a Swiss midland river: Integrated analysis of passive samplers, wild and caged fish, and vitellogenin mRNA and protein. Environ Toxicol Chem 25: 20772086.
  • 117
    Tapie N, Devier MH, Soulier C, Creusot N, Le Menach K, Ait-Aissa S, Vrana B, Budzinski H. 2011. Passive samplers for chemical substance monitoring and associated toxicity assessment in water. Water Sci Technol 63: 24182426.
  • 118
    Vermeirssen ELM, Korner O, Schonenberger R, Suter MJF, Burkhardt-Holm P. 2005. Characterization of environmental estrogens in river water using a three pronged approach: Active and passive water sampling and the analysis of accumulated estrogens in the bile of caged fish. Environ Sci Technol 39: 81918198.
  • 119
    Vermeirssen ELM, Eggen RIL, Escher BI, Suter MJF. 2008. Estrogens in Swiss rivers and effluents—Sampling matters. Chimia 62: 389394.
  • 120
    Vermeirssen ELM, Suter MJ-F, Burkhardt-Holm P. 2006. Estrogenicity patterns in the Swiss midland river Lützelmurg in relation to treated domestic sewage effluent discharges and hydrology. Environ Toxicol Chem 25: 24132422.
  • 121
    Pesce S, Morin S, Lissalde S, Montuelle B, Mazzella N. 2011. Combining polar organic chemical integrative samplers (POCIS) with toxicity testing to evaluate pesticide mixture effects on natural phototrophic biofilms. Environ Pollut 159: 735741.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CALIBRATION METHODS USED FOR POCIS
  5. FACTORS AFFECTING POCIS SAMPLING RATES
  6. CORRECTION OF SAMPLING RATES FOR EXPOSURE CONDITIONS
  7. REPORTED APPLICATIONS AND USE
  8. CONCLUSIONS AND RECOMMENDATIONS
  9. SUPPLEMENTAL DATA
  10. Acknowledgements
  11. REFERENCES
  12. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
etc_2011_sm_SupplInfo.pdf343KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.