Grouping of PFAS for human health risk assessment: Findings from an independent panel of experts.

An expert panel was convened to provide insight and guidance on per- and polyfluoroalkyl substances (PFAS) grouping for the purposes of protecting human health from drinking water exposures, and how risks to PFAS mixtures should be assessed. These questions were addressed through multiple rounds of blind, independent responses to charge questions, and review and comments on co-panelists responses. The experts agreed that the lack of consistent interpretations of human health risk for well-studied PFAS and the lack of information for the vast majority of PFAS present significant challenges for any mixtures risk assessment approach. Most experts agreed that "all PFAS" should not be grouped together, persistence alone is not sufficient for grouping PFAS for the purposes of assessing human health risk, and that the definition of appropriate subgroups can only be defined on a case-by-case manner. Most panelists agreed that it is inappropriate to assume equal toxicity/potency across the diverse class of PFAS. A tiered approach combining multiple lines of evidence was presented as a possible viable means for addressing PFAS that lack analytical and/or toxicological studies. Most PFAS risk assessments will need to employ assumptions that are more likely to overestimate risk than to underestimate risk, given the choice of assumptions regarding dose-response model, uncertainty factors, and exposure information.

Global transport of perfluoroalkyl acids via sea spray aerosol.

Perfluoroalkyl acids (PFAAs) are persistent organic pollutants found throughout the world's oceans. Previous research suggests that long-range atmospheric transport of these substances may be substantial. However, it remains unclear what the main sources of PFAAs to the atmosphere are. We have used a laboratory sea spray chamber to study water-to-air transfer of 11 PFAAs via sea spray aerosol (SSA). We observed significant enrichment of all PFAAs relative to sodium in the SSA generated. The highest enrichment was observed in aerosols with aerodynamic diameter < 1.6 µm, which had aerosol PFAA concentrations up to ∼62 000 times higher than the PFAA water concentrations in the chamber. In surface microlayer samples collected from the sea spray chamber, the enrichment of the substances investigated was orders of magnitude smaller than the enrichment observed in the aerosols. In experiments with mixtures of structural isomers, a lower contribution of branched PFAA isomers was observed in the surface microlayer compared to the bulk water. However, no clear trend was observed in the comparison of structural isomers in SSA and bulk water. Using the measured enrichment factors of perfluorooctanoic acid and perfluorooctane sulfonic acid versus sodium we have estimated global annual emissions of these substances to the atmosphere via SSA as well as their global annual deposition to land areas. Our experiments suggest that SSA may currently be an important source of these substances to the atmosphere and, over certain areas, to terrestrial environments.

Exposure and ecotoxicological risk assessment of mixtures of top prescribed pharmaceuticals in Swedish freshwaters.

Surface water concentrations of 54 pharmaceuticals were predicted for seven major Swedish rivers and the Stockholm City area basins using the STREAM-EU model. These surface water concentrations were used to predict the ecotoxicological impact resulting from the exposure of aquatic organisms to this mixture of 54 pharmaceuticals. STREAM-EU model results indicated that <10 substances were present at median annual water concentrations greater than 10 ng/L with highest concentrations occurring mostly in the more densely populated area of the capital city, Stockholm. There was considerable spatial and temporal variability in the model predictions (1-3 orders of magnitude) due to natural variability (e.g. hydrology, temperature), variations in emissions and uncertainty sources. Local mixture ecotoxicological pressures based on acute EC50 data as well as on chronic NOEC data, expressed as multi-substance potentially affected fraction of species (msPAF), were quantified in 114 separate locations in the waterbodies. It was estimated that 5% of the exposed aquatic species would experience exposure at or above their acute EC50 concentrations (so-called acute hazardous concentration for 5% of species, or aHC5) at only 7% of the locations analyzed (8 out of 114 locations). For the evaluation based on chronic NOEC concentrations, the chronic HC5 (cHC5) is exceeded at 27% of the locations. The acute mixture toxic pressure was estimated to be predominantly caused by only three substances in all waterbodies: Furosemide, Tramadol and Ibuprofen. A similar evaluation of chronic toxic pressure evaluation logically demonstrates that more substances play a significant role in causing a higher chronic toxic pressure at more sites as compared to the acute toxic pressure evaluation. In addition to the three substances contributing most to acute effects, the chronic effects are predominantly caused by another five substances: paracetamol, diclofenac, ethinylestradiol, erythromycin and ciprofloxacin. This study provides regulatory authorities and companies responsible for water quality valuable information for targeting remediation measures and monitoring on a substance and location basis.

Facing the rain after the phase out: Performance evaluation of alternative fluorinated and non-fluorinated durable water repellents for outdoor fabrics.

Fluorinated durable water repellent (DWR) agents are used to obtain water and stain repellent textiles. Due to the on-going phase-out of DWRs based on side-chain fluorinated polymers (SFP) with "long" perfluoroalkyl chains, the textile industry lacks suitable alternatives with comparable material characteristics. The constant development and optimization of SFPs for textile applications initiated more than half a century ago has resulted in a robust and very efficient DWR-technology and textiles with exceptional hydro- and oleo-phobic properties. The industry is now in the predicament that the long-chain SFPs with the best technical performance have undesirable toxicological and environmental behaviour. This study provides a comprehensive overview of the technical performance of presently available fluorinated and non-fluorinated DWRs as part of a chemical alternatives assessment (CAA). The results are based on a study with synthetic outdoor fabrics treated with alternative DWRs and tested for repellency using industrial standard and complementary methods. Using this approach, the complex structure-property relationships of DWR-polymers could be explained on a molecular level. Both short-chain SFPs and non-fluorinated DWRs showed excellent water repellency and durability in some cases while short-chain SFPs were the more robust of the alternatives to long-chain SFPs. A strong decline in oil repellency and durability with perfluoroalkyl chain length was shown for SFP DWRs. Non-fluorinated alternatives were unable to repel oil, which might limit their potential for substitution in textile application that require repellency towards non-polar liquids.

Model-predicted occurrence of multiple pharmaceuticals in Swedish surface waters and their flushing to the Baltic Sea.

An exposure assessment for multiple pharmaceuticals in Swedish surface waters was made using the STREAM-EU model. Results indicate that Metformin (27 ton/y), Paracetamol (6.9 ton/y) and Ibuprofen (2.33 ton/y) were the drugs with higher amounts reaching the Baltic Sea in 2011. 35 of the studied substances had more than 1 kg/y of predicted flush to the sea. Exposure potential given by the ratio amount of the drug exported to the sea/amount emitted to the environment was higher than 50% for 7 drugs (Piperacillin, Lorazepam, Metformin, Hydroxycarbamide, Hydrochlorothiazide, Furosemide and Cetirizine), implying that a high proportion of them will reach the sea, and below 10% for 27 drugs, implying high catchment attenuation. Exposure potentials were found to be dependent of persistency and hydrophobicity of the drugs. Chemicals with Log D > 2 had exposure potentials <10% regardless of their persistence. Chemicals with Log D  <  -2 had exposure potentials >35% with higher ratios typically achieved for longer half-lives. For Stockholm urban area, 17 of the 54 pharmaceuticals studied had calculated concentrations higher than 10 ng/L. Model agreement with monitored values had an r2 = 0.62 for predicted concentrations and an r2 = 0.95 for predicted disposed amounts to sea.

Relationships between estimated flame retardant emissions and levels in indoor air and house dust.

A significant number of consumer goods and building materials can act as emission sources of flame retardants (FRs) in the indoor environment. We investigate the relationship between the emission source strength and the levels of 19 brominated flame retardants (BFRs) and seven organophosphate flame retardants (OPFRs) in air and dust collected in 38 indoor microenvironments in Norway. We use modeling methods to back-calculate emission rates from indoor air and dust measurements and identify possible indications of an emission-to-dust pathway. Experimentally based emission estimates provide a satisfactory indication of the relative emission strength of indoor sources. Modeling results indicate an up to two orders of magnitude enhanced emission strength for OPFRs (median emission rates of 0.083 and 0.41 µg h-1 for air-based and dust-based estimates) compared to BFRs (0.52 and 0.37 ng h-1 median emission rates). A consistent emission-to-dust signal, defined as higher dust-based than air-based emission estimates, was identified for four of the seven OPFRs, but only for one of the 19 BFRs. It is concluded, however, that uncertainty in model input parameters could potentially lead to the false identification of an emission-to-dust signal.

Evaluation of human pharmaceutical emissions and concentrations in Swedish river basins.

An emissions inventory for top consumed human pharmaceuticals in Sweden was done based on national consumption data, human metabolic rates and wastewater treatment removal rates. Concentrations of pharmaceuticals in surface waters in Swedish river basins were predicted using estimated emissions from the inventory and river discharges. Our findings indicate that the top ten emitted pharmaceuticals in our study set of 54 substances are all emitted in amounts above 0.5ton/y to both surface waters and soils. The highest emissions to water were in decreasing order for Metformin, Furosemide, Gabapentin, Atenolol and Tramadol. Predicted emissions to soils calculated with the knowledge that in Sweden sludge is mostly disposed to soil, point to the highest emissions among the studied drugs coming from, in decreasing order, Metformin, Paracetamol, Ibuprofen, Gabapentin and Atenolol. Surface water concentrations in Sweden's largest rivers, all located in low density population zones, were found to be below 10ng/L for all substances studied. In contrast, concentrations in surface waters in Stockholm's metropolitan area, the most populous in Sweden, surpassed 100ng/L for four substances: Atenolol, Metformin, Furosemide and Gabapentin.

Europe-wide estuarine export and surface water concentrations of PFOS and PFOA.

The STREAM-EU model was used to predict the water concentrations, estuarine export and retention of perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) in the eleven most populated European river catchments to provide a European-wide perspective on the contamination by these substances. Emissions of PFOS and PFOA to those catchments were calculated based on population, wealth and wastewater treatment plant (WWTP) coverage and efficiency using a previously published method and used as model input. Our estimated emissions showed the lowest values for the Thames catchment (PFOS: 0.4 ton/y; PFOA: 0.2 ton/y) and the highest values for the Rhine for PFOS (1.6 ton/y) and for the Dnieper for PFOA (1.7 ton/y). The model predicted concentrations agreed reasonable well with the existing range of measurements, apart from for PFOA in the River Po, where there is a known historical industrial contamination, and PFOS in the Rhone River, where results were much higher than the few measurements available. It was concerning that the model predicted that the surface water EQS for PFOS (0.65 ng/L) was exceeded by a wide margin in all the eleven studied European river catchments. The total calculated riverine export to the seas from the eleven catchments was 4.5 ton/y of PFOS and 3.7 ton/y of PFOA with highest exported quantities from the Rhine (PFOS: 1.0 ton/y; PFOA: 1.0 ton/y) and Danube estuaries (PFOS: 0.9 ton/y; PFOA: 0.7 ton/y). For the seas where the rivers discharge, riverine discharge of PFOS was estimated to be 2.5-30 times more important as an input than atmospheric deposition, whereas for PFOA the opposite was true (atmospheric deposition was 2-10 times more important) except for very small seas.

Properties, performance and associated hazards of state-of-the-art durable water repellent (DWR) chemistry for textile finishing.

Following the phase-out of long-chain per- and polyfluoroalkyl substances (PFASs), the textile industry had to find alternatives for side-chain fluorinated polymer based durable water repellent (DWR) chemistries that incorporated long perfluoroalkyl side chains. This phase-out and subsequent substitution with alternatives has resulted in a market where both fluorinated and non-fluorinated DWRs are available. These DWR alternatives can be divided into four broad groups that reflect their basic chemistry: side-chain fluorinated polymers, silicones, hydrocarbons and other chemistries (includes dendrimer and inorganic nanoparticle chemistries). In this critical review, the alternative DWRs are assessed with regards to their structural properties and connected performance, loss and degradation processes resulting in diffuse environmental emissions, and hazard profiles for selected emitted substances. Our review shows that there are large differences in performance between the alternative DWRs, most importantly the lack of oil repellence of non-fluorinated alternatives. It also shows that for all alternatives, impurities and/or degradation products of the DWR chemistries are diffusively emitted to the environment. Our hazard ranking suggests that hydrocarbon based DWR is the most environmentally benign, followed by silicone and side-chain fluorinated polymer-based DWR chemistries. Industrial commitments to reduce the levels of impurities in silicone based and side-chain fluorinated polymer based DWR formulations will lower the actual risks. There is a lack of information on the hazards associated with DWRs, in particular for the dendrimer and inorganic nanoparticle chemistries, and these data gaps must be filled. Until environmentally safe alternatives, which provide the required performance, are available our recommendation is to choose DWR chemistry on a case-by-case basis, always weighing the benefits connected to increased performance against the risks to the environment and human health.

A large-scale model for simulating the fate & transport of organic contaminants in river basins.

We present STREAM-EU (Spatially and Temporally Resolved Exposure Assessment Model for EUropean basins), a novel dynamic mass balance model for predicting the environmental fate of organic contaminants in river basins. STREAM-EU goes beyond the current state-of-the-science in that it can simulate spatially and temporally-resolved contaminant concentrations in all relevant environmental media (surface water, groundwater, snow, soil and sediments) at the river basin scale. The model can currently be applied to multiple organic contaminants in any river basin in Europe, but the model framework is adaptable to any river basin in any continent. We simulate the environmental fate of perfluoroctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) in the Danube River basin and compare model predictions to recent monitoring data. The model predicts PFOS and PFOA concentrations that agree well with measured concentrations for large stretches of the river. Disagreements between the model predictions and measurements in some river sections are shown to be useful indicators of unknown contamination sources to the river basin.

Estimating emissions of PFOS and PFOA to the Danube River catchment and evaluating them using a catchment-scale chemical transport and fate model.

Novel approaches for estimating the emissions of perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) to surface waters are explored. The Danube River catchment is used to investigate emissions contributing to riverine loads of PFOS and PFOA and to verify the accuracy of estimates using a catchment-scale dynamic fugacity-based chemical transport and fate model (STREAM-EU; Spatially and Temporally Resolved Exposure Assessment Model for European basins). Model accuracy evaluation performed by comparing STREAM-EU predicted concentrations and monitoring data for the Danube and its tributaries shows that the best estimates for PFOS and PFOA emissions in the Danube region are obtained by considering the combined contributions of human population, wealth (based on local gross domestic product (GDP)) and wastewater treatment. Human population alone cannot explain the levels of PFOS and PFOA found in the Danube catchment waters. Introducing wealth distribution information in the form of local GDPs improves emission estimates markedly, likely by better representing emissions resulting from consumer trends, industrial and commercial sources. For compounds such as PFOS and PFOA, whose main sink and transport media is the aquatic compartment, a major source to freshwater are wastewater treatment plants. Introducing wastewater treatment information in the emission estimations also further improves emission estimates.

Air-surface exchange of polybrominated diphenyl ethers and polychlorinated biphenyls.

Air and leaf-litter samples were collected from a rural site in southern Ontario under meteorologically stable conditions in the early spring, prior to bud burst, over a three-day period to measure the simultaneous diurnal variations in polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs). PBDEs are used in a wide range of commercial products as flame retardants and are being assessed internationally as potential persistent organic pollutants. Total PBDE concentrations in the air ranged between 88 and 1250 pg m(-3), and were dominated primarily by the lighter congeners PBDEs 17, 28, and 47, and concentrations of total PCBs ranged between 96 and 950 pg m(-3), and were dominated by the lower chlorinated (tri- to tetra-) congeners. Slopes of Clausius-Clapeyron plots indicate that both PCBs and PBDEs are experiencing active air-surface exchange. Fugacities were estimated from concentrations in the air and leaf-litter and suggest near equilibrium conditions. Following the three-day intensive sampling period, 40 air samples were collected at 24-hour intervals in an attempt to evaluate the effect of bud burst on atmospheric concentrations. Total PBDE concentrations in the daily air samples ranged between 10 and 230 pg m(-3), and were dominated by the lighter congeners PBDE 17, 28, and 47, whereas concentrations of total PCBs ranged between 30 and 450 pg m(-3) during this period. It is hypothesized thatthe high PBDE concentrations observed at the beginning of the sampling period are the result of an "early spring pulse" in which PBDEs deposited in the snowpack over the winter are released with snowmelt, resulting in elevated concentrations in the surface and air. Later in the sampling period, following bud burst, PBDE concentrations in air fell to 10 to 20 pg m(-3), possibly due to the high sorption capacity of this freshly emerging foliage compartment.

Strategies for including vegetation compartments in multimedia models.

The incentives for including vegetation compartments in multimedia Level I, II and III fugacity calculations are discussed and equations and parameters for undertaking the calculations suggested. Model outputs with and without vegetation compartments are compared for 12 non-ionic organic chemicals with a wide variety of physical-chemical properties. Inclusion of vegetation compartments is shown to have a significant effect on two classes of chemicals: (1) those that are taken up by atmospheric deposition and (2) those that are taken up by transpiration through the plant roots. It is suggested that uptake from the atmosphere is important for chemicals with logK(OA) greater than 6 and a logK(AW) of greater than -6. Plant uptake by transpiration is important for chemicals with logK(OW) less than 2.5 and a logK(AW) of less than -1. At logK(OA) > 9 atmospheric uptake is dominated by particle-bound deposition and the importance of partitioning to vegetation is largely dependent on the relative magnitude of the particle deposition velocities to soil and vegetation. These property ranges can be used to determine if a chemical will significantly partition to vegetation. If the chemical falls outside the property ranges of the two classes it will probably be unnecessary to include vegetation in models for assessing environmental fate. The amount of chemical predicted to partition to vegetation compartments in the model is shown to be highly sensitive to certain model assumptions. Further experimental research is recommended to obtain more reliable equations describing equilibrium partitioning and uptake/depuration kinetics.

Gas--particle partitioning of organic compounds and its interpretation using relative solubilities.

A review is presented of the current understanding of the partitioning phenomena of relatively low vapor pressure organic chemicals between the gas phase and atmospheric particulates. The phenomena have been interpreted as adsorption, absorption, or a combination of both and corresponding theoretical equations suggested to quantify partitioning. Empirical correlations involving the chemical's vapor pressure and octanol--air partition coefficient are discussed. It is suggested that insights into selecting preferred correlations and into the nature of the partitioning phenomena can be enhanced by treating the partition coefficient as a ratio of a solubility or pseudo-solubility of the chemical in the aerosol particle to that in air. Such particle solubilities when calculated for PCBs, alkanes, and PAHs show remarkable constancy and are generally consistent with near-ideal absorption into organic matter. An exception occurs when the PAH is generated simultaneously with the aerosol and unusually high solubilities are observed, indicative of adsorption to active carbon surfaces. Recommendations are made for interpreting experimental partitioning data and for the use of correlations for predictive purposes.

Correlating the physical-chemical properties of phthalate esters using the 'three solubility' approach.

A quantitative structure-property relationship (QSPR) method for the correlation of physical-chemical properties and partition coefficients, namely the 'three solubility' approach, is described and applied to a group of 22 phthalate esters. The solubilities or 'apparent-solubilities' of these substances in the liquid state are compiled and correlated against Le Bas molar volume in the three primary media of air, water and octanol. From these solubilities the air-water (K(AW)), octanol-water (K(OW)) and octanol-air (K(OA)) partition coefficients are deduced. Estimated solubilities in water and octanol-water partition coefficients are shown to compare favourably with more recent accurate measurements. A set of selected values is presented, with error limits, which is recommended for use in modelling and assessment studies. Some environmental implications are discussed of the large range in property values for this series.

Measured and predicted volatilisation fluxes of PCBs from contaminated sludge-amended soils.

A laboratory experiment was carried out to measure volatilisation fluxes of polychlorinated biphenyls (PCBs) from sewage sludge-amended soils. The most commonly practised methods of applying sludge to agricultural land in the UK, namely, surface application, ploughing in to soil and subsurface injection, were simulated inside glass experimental chambers using an anaerobically digested sludge and a sandy loam soil. Humidified air was blown over the surface of the soil/sludge in the chambers for a period of 32 days, in order to sample a sufficient air volume to detect the volatilising PCBs. The resulting PCB volatilisation fluxes from the different sludge application methods were quantified and compared. Volatilisation fluxes of individual congeners were generally highest for the surface sludge (1-cm depth) application and slightly lower for the plough layer (5-cm depth) application. Fluxes from the subsurface layer of sludge (5-cm depth) were only quantified for the lightest congeners near to the end of the experimental run-time. Results from a multiple regression analysis showed that volatilisation fluxes of PCBs from the surface application are highly dependent on both the sludge concentration and the log of the octanol-air partition coefficient (K(OA)). A well-known soil volatilisation model, developed by Jury et al., was adapted and used to predict fluxes for the different sludge application methods during the experiment. The model predicted volatilisation fluxes that were reasonably comparable to measured fluxes for some congeners, but for others predicted fluxes that were more than an order of magnitude lower than measured fluxes. The model predicted similar loss kinetics to those observed in the experiment. Possible reasons for the dissimilarity between measured and predicted fluxes include inaccuracies in model input parameters and the fact that the models were not developed for predicting fluxes from sludge-amended soils.