Abstract
Air samples were collected from 4 field sites in Europe: 2 sites from the UK, Hazelrigg (semi-rural) and Manchester (urban); 1 site from Ireland: Mace Head (rural); and 1 site from Norway: Kjeller (rural). Additionally, air samples were taken from indoor locations in Tromsø, Norway. Air samples were collected using high-volume air samplers employing sampling modules containing glass-fibre filters (GFFs, particle phase), and glass columns with a polyurethane foam (PUF)–XAD-2–PUF sandwich (gaseous phase). Typical outdoor air volumes required for the determination of per- and polyfluorinated alkyl substances (PFAS) ranged from 500–1800 m3. GFFs and PUF–XAD columns were analysed separately to obtain information on phase partitioning. All air samples were analysed for volatile, neutral PFAS, with selected GFF samples halved for analysis of both neutral and airborne particle-bound ionic PFAS. Volatile PFAS were extracted from air samples by cold-column immersion with ethyl acetate, and were analysed by gas chromatography–mass spectrometry in the positive chemical ionisation mode (GC–PCI–MS). Ionic PFAS were extracted from GFFs by sonication in methanol, and were analysed by liquid chromatography–time-of-flight–mass spectrometry (LC–TOF–MS) using electrospray ionisation in the negative ion mode (ESI−). Perfluorooctanoate (PFOA) was often the predominant analyte found in the particulate phase at concentrations ranging from 1–818 pg m−3, and 8:2 fluorotelomer alcohol (FTOH) and 6:2 FTOH were the prevailing analytes found in the gas phase, at 5–243 pg m−3 and 5–189 pg m−3, respectively. These three PFAS were ubiquitous in air samples. Many other PFAS, both neutral and ionic, were also present, and levels of individual analytes were in the 1–125 pg m−3 range. Levels of some PFAS exceeded those of traditional persistent organic pollutants (POPs). In this study, the presence of 12:2 FTOH and fluorotelomer olefins (FTolefins), and ionic PFAS other than perfluorooctane sulfonate (PFOS) and PFOA, are reported in air samples for the first time. Concentrations of neutral PFAS were several orders of magnitude higher in indoor air than outdoor air, making homes a likely important diffuse source of PFAS to the atmosphere. Our repeated findings of non-volatile ionic PFAS in air samples raises the possibility that they might directly undergo significant atmospheric transport on particles away from source regions, and more atmospheric measurements of ionic PFAS are strongly recommended.