Kees Booij

Passive sampling in surface waters is an important method in the monitoring and risk assessment of hydrophobic and hydrophilic organic chemicals. Sampler designs can be optimized for fast equilibrium attainment (equilibrium sampling) or improved time-integrative capability (kinetic sampling). We argue that both equilibrium and kinetic sampling can be applied when aqueous concentrations do not vary with time, whereas kinetic sampling also yields useful results for time-variable concentrations that are often observed with hydrophilic compounds in surface waters. We show that these methods have similar accuracy in principle, but that their dominant error sources are different: sampler-water sorption coefficients for equilibrium sampling and sampling rates for kinetic sampling. In contrast to passive sampling of hydrophobic compounds, passive sampling of hydrophilic compounds is not as well-established, but major progress has been made over the last decade in the modeling of transport through the water boundary layer, membrane, and sorbent, while less progress has been made for transport in the biofouling layer. We recommend a more extensive use of diffusion cells as a research tool to gain a better understanding of transport through the respective subphases, leading to a greater maturity of passive water sampling of hydrophilic compounds.

Passive sampling is a crucial method for evaluating concentrations of hydrophilic organic compounds in the aquatic environment, but it is insufficiently understood to what extent passive samplers capture the intermittent emissions that frequently occur for this group of compounds. In the present study, silicone sheets and styrene-divinyl benzene-reversed phase sulfonated extraction disks with and without a polyethersulfone membrane were exposed under semi-field conditions in a 31 m³ flume at three different flow velocities. Natural processes and spiking/dilution measures caused aqueous concentrations to vary strongly with time. The data were analyzed using two analytical models that account for these time-variable concentrations: a sampling rate model and a diffusion model. The diffusion model generally gave a better fit of the data than the sampling rate model, but the difference in residual errors was quite small (median errors of 19 vs. 25% for silicone and 22 vs. 25% for SDB-RPS samplers). The sampling rate model was therefore adequate enough to evaluate the time-integrative capabilities of the samplers. Sampler performance was best for SDB-RPS samplers with a polyethersulfone membrane, despite the occurrence of lag times for some compounds (0.1 to 0.4 days). Sampling rates for this design also spanned a narrower range (80 to 110 mL/day) than SDB-RPS samplers without a membrane (100 to 660 mL/day). The effect of biofouling was similar for all compounds and was consistent with a biofouling layer thickness of 150 µm.

The quality of chemical analysis is an important aspect of passive sampling-based environmental assessments. The present study reports on a proficiency testing program for the chemical analysis of hydrophobic organic compounds in silicone and low-density polyethylene (LDPE) passive samplers and hydrophilic compounds in polar organic chemical integrative samplers. The median between-laboratory coefficients of variation (CVs) of hydrophobic compound concentrations in the polymer phase were 33% (silicone) and 38% (LDPE), similar to the CVs obtained in four earlier rounds of this program. The median CV over all rounds was 32%. Much higher variabilities were observed for hydrophilic compound concentrations in the sorbent: 50% for the untransformed data and a factor of 1.6 after log transformation. Limiting the data to the best performing laboratories did not result in less variability. Data quality for hydrophilic compounds was only weakly related to the use of structurally identical internal standards and was unrelated to the choice of extraction solvent and extraction time. Standard deviations of the aqueous concentration estimates for hydrophobic compound sampling by the best performing laboratories were 0.21 log units for silicone and 0.27 log units for LDPE (factors of 1.6 to 1.9). The implications are that proficiency testing programs may give more realistic estimates of uncertainties in chemical analysis than within-laboratory quality control programs and that these high uncertainties should be taken into account in environmental assessments.

Passive sampling devices (PSDs) offer key benefits for monitoring chemical water quality, but the uptake process of PSDs for hydrophilic compounds still needs to be better understood. Determining mass transfer coefficients of the water boundary layer (k_w) during calibration experiments and in situ monitoring would contribute toward achieving this; it allows for combining calibration data obtained at different temperature and hydrodynamic conditions and facilitate the translation of laboratory-derived calibration data to field exposure. This study compared two k_w measurement methods applied to extraction disk housings (Chemcatcher), namely, alabaster dissolution and dissipation of performance reference compounds (PRCs) from silicone. Alabaster- and PRC-based k_w were measured at four flow velocities (5–40 cm s^–1) and two temperatures (11 and 20 °C) in a channel system. Data were compared using a relationship based on Sherwood, Reynolds, and Schmidt numbers. Good agreement was observed between data obtained at both temperatures, and for the two methods. Data were well explained by a model for mass transfer to a flat plate under laminar flow. It was slightly adapted to provide a semi-empirical model accounting for the effects of housing design on hydrodynamics. The use of PRC-spiked silicone to obtain in situ integrative k_w for Chemcatcher-type PSDs is also discussed.

Exchange kinetics of organic compounds between passive samplers and water can be partly or completely controlled by transport in the sorbent. In such cases diffusion models are needed. A model is discussed that is based on a series of cosines (space) and exponentials (time). The model applies to mixed rate control by sorbent and water boundary layer under conditions of fixed aqueous concentrations (open systems, infinite water volumes, in situ sampling) and fixed amounts (closed systems, finite water volumes, ex situ sampling). Details on the implementation of the model in computational software and spreadsheet programs are discussed, including numerical accuracy. Key parameters are Biot number (ratio of internal/external transfer resistance) and sorbent/water phase ratio. Small Biot numbers are always indicative of rate control by the water boundary layer, but for large Biot numbers this may still be the case over short time scales. Application to environmental monitoring of nonpolar compounds showed that diffusion models are rarely needed for sampling with commonly used single-phase polymers. For determining sorption coefficients in batch incubations, the model demonstrated a profound effect of sorbent/water phase ratio on time to equilibrium. Application of the model to sampling of polar organic compounds by extraction disks with or without a membrane showed that moderate to major sorbent-controlled kinetics is likely to occur. This implies that the use of sampling rate models for such samplers needs to be reconsidered. Environ Toxicol Chem 2021;40:1241–1254. © 2021 SETAC

The use of single-phase passive samplers is a common method for sampling bioavailable concentrations of hydrophobic aquatic pollutants. Often such samplers are used in the kinetic stage, and in situ calibration is necessary. Most commonly, exchange kinetics are derived from the release rates of performance reference compounds (PRCs). In this study, a complementary calibration approach was developed, in which measuring the contaminant mass ratio (CMR) from two samplers with different thicknesses allows the dissolved concentrations to be determined. This new CMR calibration was tested (1) in a laboratory experiment with defined and constant concentrations and (2) in the field, at a storm water retention site. Silicone passive samplers with different thicknesses were used to sample a range of dissolved polycyclic aromatic hydrocarbons. In the laboratory study, the concentrations derived from the CMR calibration were compared with those from water extraction and passive dosing and differences below a factor 2 were found. In the field study, CMR-derived concentrations were compared to those from PRC calibration. Here, differences ranged by only a factor 1 to 3 between both methods. These findings indicate that the CMR calibration can be applied as a stand-alone or complementary calibration method for kinetic passive sampling.

Field‐based atrazine sampling rates (R_s) by the polar organic chemical integrative sampler (POCIS) were measured in nine headwater streams in three years covering 5 to 6 exposure periods of 2 to 3 weeks per site per year. Rates were best in line with the model R_s = 148 mL/d, with a standard deviation of 0.17 log units (factor 1.5). POCIS canisters reduced mass transfer coefficients of the water boundary layer by a factor of 2 as measured by alabaster dissolution rates. A mechanistic model that accounts for flow and temperature effects yielded a fair estimate of the effective exchange surface area (12.5 ± 0.8 cm²). This model could only be tested for higher flow velocities because of uncertainties associated with the measurement of flow velocities < 1 cm/s. Pictures of sorbent distributions in POCIS showed that the effective exchange surface area varied with time during the exposures. Error analysis indicated that sorbent distributions and chemical analysis were minor error sources. The main conclusion of this study is that an atrazine sampling rate of 148 mL/d yields consistent results for all three years across nine headwater streams.