Sampling Rate of Polar Organic Chemical Integrative Sampler (POCIS): Influence Factors and Calibration Methods

Pharmaceutical and pesticide mixtures in a Mediterranean coastal wetland: comparison of sampling methods, ecological risks, and removal by a constructed wetland

Pharmaceuticals and pesticides can be considered hazardous compounds for Mediterranean coastal wetland ecosystems. Although many of these compounds co-occur in environmental samples, only a few studies have been dedicated to assessing the ecotoxicological risks of complex contaminant mixtures. We evaluated the occurrence of 133 pharmaceuticals and pesticides in 12 sites in a protected Mediterranean wetland, the Albufera Natural Park (ANP), based on conventional grab sampling and polar organic chemical integrative samplers (POCIS). We assessed acute and chronic ecological risks posed by these contaminant mixtures using the multi-substance Potentially Affected Fraction (msPAF) approach and investigated the capacity of a constructed wetland to reduce chemical exposure and risks. This study shows that pharmaceuticals and pesticides are widespread contaminants in the ANP, with samples containing up to 75 different compounds. POCIS samplers were found to be useful for the determination of less predictable exposure profiles of pesticides occurring at the end of the rice cultivation cycle, while POCIS and grab samples provide an accurate method to determine (semi-)continuous pharmaceutical exposure. Acute risks were identified in one sample, while chronic risks were determined in most of the collected samples, with 5-25% of aquatic species being potentially affected. The compounds that contributed to the chronic risks were azoxystrobin, ibuprofen, furosemide, caffeine, and some insecticides (diazinon, imidacloprid, and acetamiprid). The evaluated constructed wetland reduced contaminant loads by 45-73% and reduced the faction of species affected from 25 to 6%. Our study highlights the need of addressing contaminant mixture effects in Mediterranean wetlands and supports the use of constructed wetlands to reduce contaminant loads and risks in areas with high anthropogenic pressure.

Factors affecting diffusion of polar organic compounds in agarose hydrogel applied to control mass transfer in passive samplers

Diffusive hydrogel-based passive sampler (HPS) based on diffusive gradients in thin films (DGT) is designed for monitoring polar organic contaminants in the aquatic environment. DGT technique controls the compound's overall uptake rate by adding a hydrogel layer of known thickness, which minimizes the importance of the resistive water boundary layer in the compound uptake process. In this work, we investigated several factors which may influence the diffusion of a range of aquatic contaminants in 1.5% agarose hydrogel. Diffusion in hydrogel was tested using the sheet stacking method. We demonstrated that a thin nylon netting incorporated into the diffusive hydrogel for mechanical strengthening does not significantly affect the diffusion of 11 perfluoroalkyl compounds. Further, we investigated the effect of pH in the range from 3 to 11 on the diffusion of a range of 39 aromatic amines (AAs) -36 aromatic, 2 aliphatic, and azobenzene in hydrogel. AAs were chosen as representatives of compounds with pH-dependent dissociation in water. Analysis of variance showed no significant difference in mean diffusion coefficient log D value at five pH values. The demonstration that the diffusion coefficient D and thus the sampling rate Rs are independent on pH simplifies the interpretation of data from field studies because we can neglect the influence of pH on the Rs. log D values (m2 s-1) of tested AAs ranged from to - 9.77 for 3,3'-dimethylbenzidine to - 9.19 for azobenzene. A negative correlation of log D with molar mass (log M) and molecular volume (log Vm) was observed (R =  - 0.57 and - 0.56, respectively). The diffusion coefficient presents a critical parameter for the sampling rate estimation of HPS. Theoretical sampling rates Rs of AAs were calculated for a HPS using the average D values. Theoretical Rs values calculated for AAs at 22°C ranged from 29 mL day-1 for 3,3'-dimethylbenzidine to 106 mL day-1 for 2-aminopyridine. Our calculated values of Rs are in the same range as those already published for a range of low-molecular polar organic contaminants, which supports the possibility of deriving sampler performance parameters in the field from laboratory-derived diffusivity of analytes in hydrogel.

Impacts of wastewater effluents and seasonal trends on levels of antipsychotic pharmaceuticals in water and sediments from two cold-region rivers

Most pharmaceuticals are found at trace concentrations in aquatic systems, but their continuous release and potential accumulation can lead to adverse health effects in exposed organisms. Concentrations can vary temporally, driven by variations in discharges of receiving waters, sorption to sediments, and other biotic and abiotic exchange processes. The principal aim of this research was to better understand the occurrence, trends, and dynamics of pharmaceuticals in a cold-climate, riverine environment. To this end, a suite of seven representative antipsychotic pharmaceuticals was measured upstream and downstream of two wastewater treatment plants (WWTPs) in Saskatchewan, Canada, located in the South Saskatchewan River and Wascana Creek, respectively, across three seasons. Concentrations of analytes were in the ng/L range and generally greater downstream of both WWTPs compared to upstream. Some compounds, including the tricyclic antidepressant amitriptyline, which was the most abundant analyte in water and sediment from both sites and across seasons, reached low μg/L concentrations. Data collected from this research effort indicate contamination with antipsychotic pharmaceuticals, with the potential to adversely impact exposed organisms.

A Review on Polyethersulfone Membranes in Polar Organic Chemical Integrative Samplers: Preparation, Characterization and Innovation

The membranes in polar organic chemical integrative samplers (POCIS) enclose the receiving sorbent and protect it from coming into direct contact with the environmental matrix. They have a crucial role in extending the kinetic regime of contaminant uptake, by slowing down their diffusion between the water phase and the receiving phase. The drive to improve passive sampling requires membranes with better design and enhanced performances. In this review, the preparation of standard polyethersulfone (PES) membranes for POCIS is presented, as well as methods to evaluate their composition, morphology, structure, and performance. Generally, only supplier-related morphological and structural data are provided, such as membrane type, thickness, surface area, and pore diameter. The issues related to the use of PES membranes in POCIS applications are exposed. Finally, alternative membranes to PES in POCIS are also discussed, although no better membrane has yet been developed. This review highlights the urge for more membrane characterization details and a better comprehension of the mechanisms which underlay their behavior and performance, to improve membrane selection and optimize passive sampler development.

Combination of different chromatographic and sampling modes for high-resolution mass spectrometric screening of organic microcontaminants in water

This study explores the combination of two sampling strategies (polar organic compounds integrative sampler (POCIS) vs. spot sampling) and four chromatographic retention modes (reversed-phase liquid chromatography (RPLC), hydrophilic interaction liquid chromatography (HILIC), mixed-mode liquid chromatography (MMLC) and supercritical fluid chromatography (SFC)) for high-resolution mass spectrometry (HRMS) screening of organic pollutants in water samples. To this end, a suspect screening approach, using iterative data-dependent tandem mass spectrometry (MS/MS) driven by a library of 3227 chemicals (including pharmaceuticals, pesticides, drugs of abuse, human metabolites, industrial chemicals and other pollutants), was employed. Results show that POCIS can afford a larger number of positive identifications as compared to spot sampling. On the other hand, the best suited retention mechanisms, in terms of identified analytes, are SFC, and followed by RPLC, MMLC and HILIC. However, the best combination (POCIS + SFC) would only allow the identification of 67% of the detected analytes. Thus, the combination of the two sampling strategies, spot and passive sampling, with two orthogonal retention mechanisms, RPLC and SFC, is proposed in order to maximize the number of analytes detected (89%). This strategy was applied to different surface water (river and estuary) samples from Galicia (NW Spain). A total of 155 compounds were detected at a confidence level 2a, from which the major class was pharmaceuticals (61%).

In situ calibration of polar organic chemical integrative sampler (POCIS) for monitoring of pharmaceuticals in surface waters

POCIS is the most widely applied passive sampler of polar organic substances, because it was one of the first commercially available samplers for that purpose on the market, but also for its applicability for a wide range of substances and conditions. Its main weakness is the variability of sampling performance with exposure conditions. In our study we took a pragmatic approach and performed in situ calibration for a set of 76 pharmaceuticals and their metabolites in five sampling campaigns in surface water, covering various temperature and flow conditions. In individual campaigns, R_S were calculated for up to 47 compounds ranging from 0.01 to 0.63 L d^-1, with the overall median value of 0.10 L d^-1. No clear changes of R_S with water temperature or discharge could be found for any of the investigated substances. The absence of correlation of experimental R_S with physical-chemical properties in combination with the lack of mechanistic understanding of compound uptake to POCIS implies that practical estimation of aqueous concentrations from uptake in POCIS depends on compound-specific experimental calibration data. Performance of POCIS was compared with grab sampling of water in seven field campaigns comprising multiple sampling sites, where sampling by both methods was done in parallel. The comparison showed that for 25 of 36 tested compounds more than 50% of POCIS-derived aqueous concentrations did not differ from median of grab sampling values more than by a factor of 2. Further, for 30 of 36 compounds, more than 80% of POCIS data did not differ from grab sampling data more than by a factor of 5. When accepting this level of accuracy, in situ derived sampling rates are sufficiently robust for application of POCIS for identification of spatial and temporal contamination trends in surface waters.