Stefan P.J. van Leeuwen

Perfluoroalkyl and polyfluoroalkyl substances in the environment: Terminology, classification, and origins

The primary aim of this article is to provide an overview of perfluoroalkyl and polyfluoroalkyl substances (PFASs) detected in the environment, wildlife, and humans, and recommend clear, specific, and descriptive terminology, names, and acronyms for PFASs. The overarching objective is to unify and harmonize communication on PFASs by offering terminology for use by the global scientific, regulatory, and industrial communities. A particular emphasis is placed on long-chain perfluoroalkyl acids, substances related to the long-chain perfluoroalkyl acids, and substances intended as alternatives to the use of the long-chain perfluoroalkyl acids or their precursors. First, we define PFASs, classify them into various families, and recommend a pragmatic set of common names and acronyms for both the families and their individual members. Terminology related to fluorinated polymers is an important aspect of our classification. Second, we provide a brief description of the 2 main production processes, electrochemical fluorination and telomerization, used for introducing perfluoroalkyl moieties into organic compounds, and we specify the types of byproducts (isomers and homologues) likely to arise in these processes. Third, we show how the principal families of PFASs are interrelated as industrial, environmental, or metabolic precursors or transformation products of one another. We pay particular attention to those PFASs that have the potential to be converted, by abiotic or biotic environmental processes or by human metabolism, into long-chain perfluoroalkyl carboxylic or sulfonic acids, which are currently the focus of regulatory action. The Supplemental Data lists 42 families and subfamilies of PFASs and 268 selected individual compounds, providing recommended names and acronyms, and structural formulas, as well as Chemical Abstracts Service registry numbers. Integr Environ Assess Manag 2011;7:513–541.

Competitive Binding of Poly- and Perfluorinated Compounds to the Thyroid Hormone Transport Protein Transthyretin

Due to their unique surfactant properties, poly- and perfluorinated compounds (PFCs) have been extensively used and can be found all over the environment. Concern about their environmental fate and toxicological properties has initiated several research projects. In the present study, we investigated if PFCs can compete with thyroxine (T(4), i.e., the transport form of thyroid hormone) for binding to the human thyroid hormone transport protein transthyretin (TTR). Such competitive capacity may lead to decreased thyroid hormone levels as previously reported for animals exposed to PFCs. Twenty-four PFCs, together with 6 structurally similar natural fatty acids, were tested for binding capacity in a radioligand-binding assay. The binding potency decreased in the order: perfluorohexane sulfonate > perfluorooctane sulfonate/perfluorooctanoic acid > perfluoroheptanoic acid > sodium perfluoro-1-octanesulfinate > perfluorononanoic acid, with TTR binding potencies 12.5-50 times lower than the natural ligand T(4). Some lower molecular weight compounds with structural similarity to these PFCs were > 100 times less potent than T(4). Simple descriptors based on the two-dimensional molecular structures of the compounds were used to visualize the chemical variation and to model the structure-activity relationship for the competitive potencies of the TTR-binding compounds. The models indicated the dependence on molecular size and functional groups but demanded a more detailed description of the chemical properties and data for validation and further quantitative structure-activity relationship (QSAR) development. Competitive binding of PFCs to TTR, as observed for human TTR in the present study, may explain altered thyroid hormone levels described for PFC-exposed rats and monkeys. Median human blood levels of the most potent TTR-binding PFCs are one to two orders of magnitude lower than concentration at 50% inhibition (IC(50)) values determined in the present study. In addition, this study contributes to the understanding of the bioaccumulation of PFCs in man and possibly in other wildlife species.

Levels of perfluorinated compounds in food and dietary intake of PFOS and PFOA in the Netherlands.

This study presents concentrations of perfluorinated compounds in food and the dietary intake of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in The Netherlands. The concentrations of perfluorinated compounds in food were analyzed in pooled samples of foodstuffs randomly purchased in several Dutch retail store chains with nation-wide coverage. The concentrations analyzed for PFOS and PFOA were used to assess the exposure to these compounds in The Netherlands. As concentrations in drinking water in The Netherlands were missing for these compounds, conservative default concentrations of 7 pg/g for PFOS and 9 pg/g for PFOA, as reported by European Food Safety Authority, were used in the exposure assessment. In food, 6 out of 14 analyzed perfluorinated compounds could be quantified in the majority of the food categories (perfluoroheptanoic acid (PFHpA), PFOA, perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), perfluoro-1-hexanesulfonate (PFHxS), and PFOS). The highest concentration of the sum of these six compounds was found in crustaceans (825 pg/g product, PFOS: 582 pg/g product) and in lean fish (481 pg/g product, PFOS: 308 pg/g product). Lower concentrations were found in beef, fatty fish, flour, butter, eggs, and cheese (concentrations between 20 and 100 pg/g product; PFOS, 29-82 pg/g product) and milk, pork, bakery products, chicken, vegetable, and industrial oils (concentration lower than 10 pg/g product; PFOS not detected). The median long-term intake for PFOS was 0.3 ng/kg bw/day and for PFOA 0.2 ng/kg bw/day. The corresponding high level intakes (99th percentile) were 0.6 and 0.5 ng/kg bw/day, respectively. These intakes were well below the tolerable daily intake values of both compounds (PFOS, 150 ng/kg bw/day; PFOA, 1500 ng/kg bw/day). The intake calculations quantified the contribution of drinking water to the PFOS and PFOA intake in The Netherlands. Important contributors of PFOA intake were vegetables/fruit and flour. Milk, beef, and lean fish were important contributors of PFOS intake.

Brominated flame retardants in fish and shellfish - levels and contribution of fish consumption to dietary exposure of Dutch citizens to HBCD.

In order to determine the contamination with brominated flame retardants (BFR) in fish regularly consumed by Dutch citizens, 44 samples of freshwater fish, marine fish, and shellfish were analyzed for polybrominated diphenyl ethers (PBDE), tetrabromobisphenol-A (TBBP-A) and its methylated derivative (me-TBBP-A), and hexabromocyclododecane (HBCD), including its alpha-, beta- and gamma-diastereomers. The highest BFR concentrations were found in pike-perch and eel from the highly industrialized and urbanized rivers Rhine and Meuse. The sum concentrations of BDE 28, 47, 99, 100, 153, 154, 183, 209, and brominated biphenyl (BB) 153 and HBCD (selection based on The European Food Safety Authority monitoring recommendation) ranged from below quantification limits to 17 ng/g wet weight (ww) in marine fish and in freshwater fish from 0.6 ng/g ww in pike-perch to 380 ng/g ww in eel. The BDE congener profile in all fish and shellfish samples is dominated by BDE 47, followed by BDE 99, except for eel in which BDE 100 is higher than BDE 99. BDE 209 was detected in two mussel samples, most likely due to BDE 209 contaminated particulate matter in their intestines. Total-HBCD (as determined by GC/electron capture negative ion (ECNI)-MS) was detected in 22 out of the 44 samples in concentrations between 0.20 ng/g in marine fish and 230 ng/g ww in eel. Three HBCD diastereomers were determined by HPLC/ESI-MS/MS. alpha-HBCD was the prevalent congener in most fish samples, followed by gamma-HBCD. beta-HBCD, TBBP-A and me-TBBP-A were only detected in a few samples and at low concentrations. A considerable difference was found between HBCD results obtained from GC/ECNI-MS and HPLC/ESI-MS/MS: the GC/ECNI-MS results were 4.4 times higher, according to regression analysis. There is hardly any data on human dietary exposure to HBCD available. We have estimated the fish-related dietary exposure of HBCD for the average Dutch population. The medium bound intake was estimated at 8.3 ng/day for a 70-kg person (0.12 ng/kg bodyweight/day). For this estimation, we relied mostly on HPLC/ESI-MS/MS data as we argue that these results are more accurate than those obtained by GC/ECNI-MS.

Dietary exposure to dioxins and dioxin-like PCBs in The Netherlands anno 2004

In this study, representative occurrence data for PCDD/Fs and dioxin-like PCBs in food were obtained and used to estimate dietary exposure of the Dutch population. Food composite samples were analyzed as well as single fish and vegetables samples. Total dioxin concentrations in animal products ranged from 0.05 pg TEQ/g product in poultry to 2.5 pg TEQ/g product (using TEF(2006)) in fish (shrimp), with 0.12pg TEQ/g product being the lowest concentrations measured in fish (tuna). In vegetable products, concentrations ranged from 0.00002 pg TEQ/g product (white kale) to 0.19 pg TEQ/g (oils and fats). A long-term dietary exposure distribution was calculated using Monte Carlo Risk Assessment software. The lower bound median exposure of the Dutch population to PCDD/Fs and dioxin-like PCBs was estimated at 0.8 pg WHO-TEQ/kgbw/d, half of which were dioxin-like PCBs. Dairy was the main source (38%) due to its high consumption. Time-trend analysis shows that the exposure to dioxins has further decreased by 35% over the past five years. This is due to lower levels of dioxin-like compounds in most of the foods, mainly influenced by lower levels in meat and milk. The use of the new TEFs gives an exposure reduction of 10% with respect to TEF(1998). Still, 4% of the Dutch population exceeds the exposure limit of 14 pg/kgbw/week as set by the EU.

Analytical challenges hamper perfluoroalkyl research

The growing concern over these organohalogens, some of which have been found in human blood and appear to be widespread in the environment, led researchers to gather in Hamburg, Germany, in 2003 to evaluate the current state of methods to analyze for the organic contaminants. Jonathan Martin of the University of Toronto and 20 colleagues from industry, government, and academia summarize the main recommendations from the workshop.

Tetrahydrofuran-water extraction, in-line clean-up and selective liquid chromatography/tandem mass spectrometry for the quantitation of perfluorinated compounds in food at the low picogram per gram level

A new solvent extraction system was developed for extraction of PFCs from food. The extraction is carried out with 75:25 (v/v) tetrahydrofuran:water, a solvent mixture that provides an appropriate balance of hydrogen bonding, dispersion and dipole-dipole interactions to efficiently extract PFCs with chains containing 4-14 carbon atoms from foods. This mixture provided recoveries above 85% from foods including vegetables, fruits, fish, meat and bread; and above 75% from cheese. Clean-up with a weak anion exchange resin and Envi-carb SPE, which were coupled in line for simplicity, was found to minimize matrix effects (viz. enhancement or suppression of electrospray ionization). The target analytes (PFCs) were resolved on a perfluorooctyl phase column that proved effective in separating mass interferences for perfluorooctane sulfonate (PFOS) in fish and meat samples. The mass spectrometer was operated in the negative electrospray ionization mode and used to record two transitions per analyte and one per mass-labeled method internal standard. The target PFCs were quantified from solvent based calibration curves. The limits of detection (LODs) were as low as 1-5 pg analyte g(-1) food; by exception, those for C(4) and C(5) PFCs were somewhat higher (25-30 pg g(-1)) owing to their less favourable mass response. To the best of our knowledge these are among the best LODs for PFCs in foods reported to date. The analysis of a variety of foods revealed contamination with PFCs at levels from 4.5 to 75 pg g(-1) in 25% of samples (fish and packaged spinach). C(10)-C(14) PFCs were found in fish, which testifies to the need to control long-chain PFCs in this type of food. The proposed method is a useful tool for the development of a large-scale database for the presence of PFCs in foods.

Recent developments in trace analysis of poly- and perfluoroalkyl substances

AbstractRecent developments, improvements, and trends in the ultra-trace determination of per- and polyfluoroalkyl substances (PFASs) in environmental and human samples are highlighted and the remaining challenges and uncertainties are outlined and discussed. Understanding the analytical implications of such things as adsorption of PFASs to surfaces, effects of differing matrices, varying PFAS isomer response factors, potential bias effects of sampling, sample preparation, and analysis is critical to measuring highly fluorinated compounds at trace levels. These intricate analytical issues and the potential consequences of ignoring to deal with them correctly are discussed and documented with examples. Isomer-specific analysis and the development of robust multi-chemical methods are identified as topical trends in method development for an ever-increasing number of PFASs of environmental and human interest. Ultimately, the state-of-the-art of current analytical method accuracy is discussed on the basis of results from interlaboratory comparison studies. FigureSeparation of the linear, mono-trifluoromethyl branched, and di-trifluoromethyl branched structural isomers of PFOS and PFOA by ultra-performance liquid chromatography using a conventional C18 reversed-phase column. The PFOS and PFOA structural isomers were detected by tandem mass spectrometry

Use of a presolvent to include volatile organic analytes in the application range of on-line solid-phase extraction–gas chromatography–mass spectrometry

The application range of the on-line solid-phase extraction–gas chromatographic (SPE–GC) analysis of aqueous samples has been extended to volatile analytes. In the new set-up, after conventional aqueous-sample loading and drying of the SPE cartridge with nitrogen gas, 30–50 μl of an organic solvent, the so-called presolvent, such as methyl acetate or ethyl acetate are introduced into the retention gap prior to the actual desorption to ensure that a solvent film is already present in the retention gap when the introduction of the analyte-containing desorption solvent starts. This procedure allows the recovery of analytes as volatile as monochlorobenzene and xylene. Aspects such as the type of retaining precolumn, and the type and amount of presolvent have been studied systematically to explain the performance of the novel set-up. Actually, when using 50 μl of presolvent, the use of a retaining precolumn did not have any significant influence on the recovery of the volatile analytes. The modified SPE–GC procedure was tested by analysing 10 ml of river Rhine water spiked at the 0.5 μg/l level with about 80 microcontaminants covering a wide range of volatility. The test compounds included chlorobenzenes, substituted and nonsubstituted aromatic compounds, anilines and phenolic compounds and organonitrogen and organophosphorus pesticides. The system performance in terms of recovery (typically 70–115% at the 0.5 μg/l level) and repeatability (R.S.D. values typically 1–9%; n=7) was satisfactory, even for monochlorobenzene, the most volatile analyte of the test mixture. Low recoveries due to early breakthrough (polar analytes) or adsorption to the tubing (apolar analytes) were observed for a few analytes only. The detection limits in SPE–GC–MS using full-scan acquisition generally were 20–50 ng/l.

Perfluoroalkyl substances in polar bear mother-cub pairs: A comparative study based on plasma levels from 1998 and 2008

Perfluoroalkyl substances (PFASs) are protein-binding blood-accumulating contaminants that may have detrimental toxicological effects on the early phases of mammalian development. To enable an evaluation of the potential health risks of PFAS exposure for polar bears (Ursus maritimus), an exposure assessment was made by examining plasma levels of PFASs in polar bear mothers in relation to their suckling cubs-of-the-year (~4 months old). Samples were collected at Svalbard in 1998 and 2008, and we investigated the between-year differences in levels of PFASs. Seven perfluorinated carboxylic acids (∑₇PFCAs: PFHpA, PFOA, PFNA, PFDA, PFUnDA, PFDoDA, and PFTrDA) and two perfluorinated sulfonic acids (∑₂PFSAs: PFHxS and PFOS) were detected in the majority of the mothers and cubs from both years. In mothers and cubs, most PFCAs were detected in higher concentrations in 2008 than in 1998. On the contrary, levels of PFOS were lower in 2008 than in 1998, while levels of PFHxS did not differ between the two sampling years. PFOS was the dominating compound in mothers and cubs both in 1998 and in 2008. Concentration of PFHpA did not differ between mothers and cubs, while concentrations of PFOA, PFNA, PFDA, PFUnDA, PFDoDA, PFTrDA, PFHxS, and PFOS were higher in mothers than in their cubs. Except from PFHpA, all compounds correlated significantly between mothers and their cubs. The mean cub to mother ratios ranged from 0.15 for PFNA to 1.69 for PFHpA. On average (mean±standard error of mean), the levels of ∑₇PFCAs and ∑₂PFSAs in cubs were 0.24±0.01 and 0.22±0.01 times the levels in their mothers, respectively. Although maternal transfer appears to be a substantial source of exposure for the cubs, the low cub to mother ratios indicate that maternal transfer of PFASs in polar bears is relatively low in comparison with hydrophobic contaminants (e.g. PCBs). Because the level of several PFASs in mothers and cubs from both sampling years exceeded the levels associated with health effects in humans, our findings raise concern on the potential health effects of PFASs in polar bears from Svalbard. Effort should be made to examine the potential health effects of PFASs in polar bears.