Pine Needles for the Screening of Perfluorinated Alkylated

Article pubs.acs.org/est

Pine Needles for the Screening of Perfluorinated Alkylated Substances (PFASs) along Ski Tracks Mária Chropeňová,† Pavlína Karásková,† Roland Kallenborn,‡ Eva Klemmová Gregušková,† and Pavel Č upr*,† †

Environ. Sci. Technol. 2016.50:9487-9496. Downloaded from pubs.acs.org by IOWA STATE UNIV on 01/19/19. For personal use only.

Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Kamenice 753/5, Pavilion A29, 625 00 Brno, Czech Republic ‡ Department of Chemistry, Biotechnology and Food Science (IKBM), Norwegian University of Life Sciences (NMBU), P.O. Box 5003, Christian M. Falsens veg 1, NO-1432 Ås, Norway S Supporting Information *

ABSTRACT: Perfluorinated alkylated substances (PFASs) are today considered persistent, toxic, and bioaccumulative contaminants. Perfluorooctansulfonate (PFOS) and perfluorooctanoic acid (PFOA) are currently listed as priority substances under the UNEP global convention for the regulation of POPs. A previous study reported higher levels of PFASs in pine needles near ski areas. Their application as stain repellents in modern outdoor clothes and in ski waxes is assumed to be a potential source. Pine trees (Pinus mugo in Slovakia and Pinus sylvestris in Norway) were chosen for sampling in ski resorts. Relative distributions, overall concentrations, trend estimates, elevation patterns, and distance from primary sources were assessed. PFOA was the predominant PFAS constituent in pine needles from Slovakia (8−93%). In Norway, the most-abundant PFAS was perfluorobutanoic acid (PFBA: 3−66%). A difference in product composition (particularly in ski waxes) and differences in Norwegian and Slovakian regulations are considered to be the primary reason for these differences. Open application of PFOA in industry and products has been banned in Norway since 2011. The replacement of PFOA with short-chain substitutes is thus considered the reason for the observed pattern differences in the analyzed pine needles. Regular monitoring and screening programs are recommended.

■

INTRODUCTION Perfluorinated alkylated substances (PFASs) are a group of high-production-volume chemicals identified as persistent environmental pollutants in virtually all environmental compartments. PFASs are characterized by a hydrophobic fluorinated carbon tail attached to a polar hydrophilic head.1 The PFAS group consists of perfluorinated sulfonic acids (PFSAs), perfluorinated carboxylic acids (PFCAs), fluorotelomer alcohols (PFTOHs), perfluorinated sulphonamides (FOSAs), and sulphonamido ethanols (FOSEs) as well as medium- and high-molecular-weight fluoropolymers.2 PFASs are characterized by extreme environmental stability. This is mainly caused by the stability of the carbon−fluorine bond. PFASs possess special physical−chemical properties that make them highly attractive in a variety of technical applications. They are highly surface-active and amphiphilic. As such, they are applied as stain repellents, surface coatings, and firefighting agents.1 However, selected PFAS are known to bioaccumulate in food webs and are even found in substantial concentrations in humans.3−5 Due to their documented harmful effects on the environment, in 2009, perfluorooctansulfonate (PFOS) was officially added as a persistent organic pollutant (POP) to the © 2016 American Chemical Society

priority list (Annex B: restriction of use) of the Stockholm Convention for the global regulation of POPs, and perfluorooctanoic acid (PFOA) is currently under consideration.6,7 PFOA is characterized as a highly persistent, bioaccumulative, and toxic (PBT) substance.8,9 Consequently, the Persistent Organic Pollutants Review Committee (POPRC) of the UNEP convention for the global regulation of persistent organic pollutants (Stockholm “POPs” Convention) confirmed that PFOA meets the official priority criteria (persistence, bioaccumulation, potential for long-range environmental transport, and adverse effects) at their last meeting in October 2015 in Rome. A working group is currently preparing a draft risk profile for PFOA.10 This illustrates the international regulative priorities for PFAS-related compounds and for PFOA and PFOS in particular. PFASs have been produced since the 1950s and are used in many industrial and household products.11−13 Received: Revised: Accepted: Published: 9487

May 13, 2016 July 25, 2016 July 26, 2016 July 26, 2016 DOI: 10.1021/acs.est.6b02264 Environ. Sci. Technol. 2016, 50, 9487−9496

Article

Environmental Science & Technology

direction of prevailing wind from the ski resorts and crosscountry tracks. For each location, two representative sites were chosen. The Slovakian sampling sites are shown in Figure 1 and the Norwegian sampling sites in Figure 2.

In addition to the above-mentioned application areas, PFASs are also used in ski waxes.14−16 Regardless of pollutant type, the potential environmental risk is usually assessed by long-term monitoring of potentially contaminated environments and a subsequent exposure assessment. However, in remote sites (such as ski areas and high mountain habitats), the installation of advanced, powerconsuming monitoring equipment is not ideal due to limited power supplies and logistical constraints (transportation and maintenance of equipment). Thus, for integrated sampling, naturally occurring, potentially accumulating sample material (i.e., suitable vegetation in terrestrial environments for metal surveys and mussels in marine environments for POPs) are often selected.17,18 For organic pollutants, pine needles have also been previously evaluated as passive samples for integrated monitoring purposes.19,20 On the basis of the positive results of these previous investigations, this study validated pine needles (Pinus species) as suitable sample material21,22 and integrated passive sampling for the screening of selected PFAS along cross-country and downhill ski tracks in Slovakia and Norway. Due to their wax cuticle, pine needles have strong adsorptive surface properties. The versatility of this sample type has been previously proven for legacy pollutants.23−25 However, pine needles have never been applied as a monitoring tool for PFAS release and exposure risks associated with skiing activities. From a logistical perspective, pine needles are ubiquitously available in most ski resort areas and are easy to collect and store for analytical purposes, as shown by a suit of studies on legacy POPs.22,25−28 A first pilot study conducted in 2014 confirmed the suitability of pine needles for PFAS screening. However, during this preliminary study, mountain pine (Pinus mugo) needles were exclusively evaluated in selected Slovakian mountain regions without a direct link to potential local sources.22 The relevance of ski waxes as a potential source of environmental contamination of PFASs has already been confirmed in earlier studies, mainly focusing on occupational aspects.29,30 Fluorinated ski waxes usually contain semifluorinated n-alkanes (SFAs) and perfluorinated compounds.31,32 This usually also includes perfluoroalkyl carboxylic acids (PFCAs). The global production of ski waxes is estimated to be several tons per year.14 As an additional source in recreational skiing areas, PFAS may also be released to the environment from outdoor clothing. In addition to direct exposure to PFASs during skiing preparation and recreational activities, PFAS emissions from outdoor clothes are also expected to contribute to the overall exposure of persons engaging in these types of recreational activities during the winter.13 Earlier studies showed that PFASs associated with aerosol and atmospheric particles are preferably deposited on the wax surface layer of needles.22 In addition, pine trees are typical vegetation in skiing areas and thus easily available as sampling material, regardless of location. The selected pine trees (dwarf mountain pine, P. mugo, in Slovakia; Scots pine, Pinus sylvestris, in Norway) were sampled in close vicinity and inside the ski resort grounds chosen for this study. In addition to PFAS level and distribution, this study examined sampling site-specific criteria and uptake profiles in the two pine species as well as the annual needle growth characteristics. Sampling Sites. Pine needles from trees in the vicinity of representative skiing locations were collected for quantitative PFAS analysis. The chosen sampling location lay in the

Figure 1. Slovakian sampling sites: 1, Martinske Hole Mountain (E); 2, Martinske Hole Mountain (W); 3, Krivan Mountain (SW); 4, Krivan Mountain (N); 5, Chopok Mountain; 6, Sivy vrch Mountain (SE); 7, Sivy vrch Mountain (E); 8, Ostredok Mountain; 9, Zilina (negative reference site); 10, Kopske sedlo Mountain (positive reference site).

In Slovakia. Kopske seddle (Site 10) is a high mountain location in the Belianske Tatra Mountains. This site is characterized by direct atmospheric pollution from nearby local sources, most probably from Polish regions. Zilina (Site 9) is a regional center with industrial activity and an urban area. However, this area is not expected to significantly influence the

Figure 2. Norwegian sampling sites: 11, Oslo Winter Park; 12, Hafjell Ski Center; 13, Birkebeineren, Lillehammer; 14, Kongsberg Ski Center; 15, Holmenkollen, Holmenkollen ski jump; 16, Holmenkollen, Midtstuen ski jump; 17, Oslo (negative reference site); 18, Ås (positive reference site). 9488

DOI: 10.1021/acs.est.6b02264 Environ. Sci. Technol. 2016, 50, 9487−9496

Article


Switzerland) in methanol with the addition of ammonium acetate (Fluka Chemie GmbH; Buchs, Germany; 400 mg L−1 methanol). Prior to extraction, the following recovery standards were added: 13C8PFOA and 13C8PFOS, 50 μL, 80 ng mL−1. The concentrated methanol extracts for PFAS analysis were diluted with 200 mL of high-performance liquid chromatography (HPLC) gradient-grade water (Fisher Scientific; Loughborough, U.K.) followed by purification using SPE cartridges (Oasis Wax 6 cm3, 500 mg, Waters; Bedford, CA), which were conditioned with 5 mL of methanol, 10 mL of basic methanol (methanol modified with 4 mL of NH4OH, ammonia solution with 4 mL L−1 methanol), and 5 mL of HPLC gradient-grade water. Diluted extracts were passed through the cartridges at one drip per second with a final wash step with wash buffer (ammonium acetate and acetic acid in water). Elution was performed using 6 mL of neutral methanol and 8 mL of basic methanol. The resulting eluents were collected in polypropylene falcon tubes (Eastport Lifescience; Praha, Czech Republic), concentrated under a gentle stream of nitrogen to almost dryness. Then, a 1 mL volume with 0.5 mL of MeOH and 0.5 mL of 5 mM ammonium acetate in water was added. The rediluted extracts were finally cleaned through a syringe filter (nylon membrane, 13 mm diameter and 0.45 μm pore size; Labicom Czech Republic) and transferred into HPLC autosampler vials. A total of eight mass-labeled internal standards (13C2 PFHxA, 13C4 PFOA, 13C5 PFNA, 13C2 PFDA, 13 C2 PFUnDA, 13C2 PFDoDA, 18O2 PFHxS, and 13C4 PFOS) (Wellington Laboratories Inc.; Guelph, Canada) were added as a final step before high-performance liquid chromatography− mass spectrometry (HPLC-MS) analysis. The separation, identification, and quantification of all target PFASs were performed using HPLC with an Agilent 1290 (Agilent Technologies; Palo Alto, California) connected to a mass spectrometer QTRAP 5500 (ABSciex; Foster City, California). Chromatographic separation was performed at 20 °C on a SYNERGI 4 μ Fusion RP 80 Å 50 mm × 2 mm column with a corresponding precolumn (SecurityGuard: C18 4 × 2 mm, Phenomenex; Torrance, CA). The flow rate was set to 200 μL min−1. Mobile phase A was methanol, and mobile phase B was methanol and 5 mM ammonium acetate in water, 55/45 (v/v). Aliquots of 10 μL were injected into the column, and gradient elution was used for PFAS separation. Before the next separation, the column was equilibrated using the initial content of the mobile phase for 3 min. The mass spectrometer was operated in electrospray negative ionization mode (ESI) using two MRM transitions for each compound (except perfluorobutanoic acid, PFBA) at 450 °C and ion voltage of 4500 V (Table S3).

skiing sites. The remaining eight sampling sites are located in the Western Carpathians region (Little Fatra: Sites 1, 2, 3, and 4; Great Fatra: Site 8; and Tatra Mountains: Sites 5, 6, and 7) and are all considered background locations with respect to PFASs. The sampling of needles from P. mugo (dwarf mountain pine) was conducted from June to July 2014 in Slovakia. This species is considered representative for the high mountain areas in Slovakia. In Norway. Ås (municipality in Akershus County) (Site 18) was chosen as a potential contaminated site due to the proximity to the Ås municipality. In addition, as an urban background site (low or minimal PFAS levels expected), the pine trees in the park close to the royal palace in Oslo were sampled as reference. All remaining six sampling sites (Sites 11, 12, 13, 14, 15, and 16) were chosen from southeastern Norwegian ski resorts. Needles of P. sylvestris (Scots pine) were sampled in October 2015. This tree is the only native Pinus species in Norway. The geographical coordinates were recorded using Google maps and are shown in Tables S1 and S2. Analyzed Compounds. All target perfluorinated alkylated substances (PFASs) are shown in Table 1. Table 1. Analyzed Perfluorinated Alkylated Substances IUPAC name

acronym

CAS number

perfluorobutanoic acid perfluoropentanoic acid perfluorohexanoic acid perfluoroheptanoic acid perfluorooctanoic acid perfluorononanoic acid perfluorodecanoic acid perfluoroundecanoic acid perfluorododecanoic acid perfluorotridecanoic acid perfluorotetradecanoic acid perfluorobutanesulfonic acid perfluorohexanesulfonate perfluoroheptanesulfonate perfluorooctanesulfonate perfluorodecanesulfonic acid

PFBA PFPA PFHxA PFHpA PFOA PFNA PFDA PFUnDA PFDoDA PFTrDA PFTeDA PFBS PFHxS PFHpS PFOS PFDS

375-22-4 2706-90-3 307-24-4 375-85-9 335-67-1 375-95-1 335-76-2 2058-94-8 307-55-1 72629-94-8 376-06-7 375-73-5 432-50-7 375-92-8 1763-23-1 39108-34-4

■

MATERIALS AND METHODS Solvents used for sample preparation and analysis were purchased from Fluka (Buchs, Germany) and were of environmental trace analytical quality. Needles were separated from the branches and sorted by age into 1 to 5 year old needles in the case of P. mugo and into 1 to 3 year old needles in the case of P. sylvestris.33 Needles were wrapped in aluminum foil and stored in polyethylene bags in a freezer at −20 °C until further processing. A brief method description is provided here. However, more details on the sampling method can be found in a previous publication.22 Needles were homogenized using an electric grinder (IKA A11 basic, Sigma-Aldrich; Praha, Czech Republic). A total of 100 mL of Milli-Q water (Super-Q Water System, Millipore Corp.; Bedford, MA) was added prior to lyophilization. For lyophilization, a standard freeze-drying unit was used (Scanvac Coolsafe TM, Trigon-plus, Č estlice; Ř ı ́cǎ ny u Prahy, Czech Republic). Preweighed 10 g samples were extracted with an automatic extractor (Büchi B-811, Labortechnik AG; Flawil,

■

QUALITY ASSURANCE AND QUALITY CONTROL The efficiency of the extraction method was validated by spiking pine needle homogenates with 50 μL of a 80 ng mL−1 mixture of native perfluorinated alkylated substances followed by the extraction method described above.22 Percentage recoveries for all native PFASs are reported in Table S4. Evaluation and quantification of data was done by internalstandard-based calculations using nine-point quadratic calibration curves (r2 > 0.99) ranging from 0.004 to 20 ng mL−1. Internal standards (13C4 PFBA, 13C2 PFHxA, 13C4 PFOA, 13C5 PFNA, 13C2 PFDA, 13C2 PFUnDA, 13C2 PFDoDA, 18O2 PFHxS, and 13C4 PFOS) were added to all samples prior to extraction, and the same amount of these standards was added to each calibration point. The reported concentrations are 9489


Article


Figure 3. Relative distribution of PFASs in 2 year old needles from Slovakian sites (Sites 1−10) and Norwegian sites (Sites 11−18).

surface-active, and amphiphilic chemical pollutants,35 significant alteration of distribution patterns due to chemical transformation for PFASs on pine needles was not expected. Therefore, the usage of different pine species (predominant in the respective countries) is considered appropriate for the performed comparison. The morphology of the needles of each species (e.g., specific leaf area and surface) may have a crucial role in the distribution of pollutants.36,37 The needles of P. mugo are densely set on the branches, 3−8 cm long, 1.5−2.2 mm wide, and dark or light green. The needles of P. sylvestris are 4−7 cm long, 1−2 mm wide, and green. The anatomy of needles from both tree species shows a thick hypodermis and several external resin ducts.38 Thus, differences in the physiology of needles between tree species were not relevant for leaf area and needle surface. Because Slovakia and Norway are in different climate zones, it is also necessary to consider the differences in weather conditions, atmospheric transport, etc. Samples were collected in spring in Slovakia and in autumn in Norway. However, the weather (temperature and precipitation) was comparable between Slovakian mountains in spring and southeastern Norway in autumn. A pair of reference sites (Sites 9 and 10) in Slovakia, where both tree species were sampled, were used to compare the overall accumulation of PFASs at the beginning of spring 2015. The sum of PFAS concentrations was 12 ng g−1 dw for P. mugo, 10 ng g−1 dw for P. sylvestris in Zilina (Site 9), and 25 ng g−1 dw for P. mugo, 18 ng g−1 dw for P. sylvestris were sampled for comparing different tree species at Kopske seddle (Site 10). It should be noted that the ability to accumulate pollutants depends on the chosen tree species. In the case of this presented study, no marked differences were observed that would prohibit the comparison of results across the chosen species. However, many factors may affect the process of sampling, analysis, and evaluation of results. Relative Distributions. Relative distribution profiles for the quantified PFASs were used for comparison of all samples. Significant differences in both distribution patterns and concentrations were found between Slovakian and Norwegian locations (Figure 3). Perfluorooctanoic acid had the highest relative abundance in all Slovakian samples (8−93%). The lowest PFOA concentrations were at the two reference sites

laboratory-blank-corrected (one blank in each batch of 10 samples). Recoveries of 13C8PFOA in real samples ranged from 65% to 118%, with an average value of 95 ± 15%, and of 13 C8PFOS from 56% to 110% with an average value of 90 ± 18%. Method quantification limits (MQLs) in ng g−1 dw were defined as the mean concentration of procedural blanks plus 10× the standard deviation of blank response. All results and MQL values are listed in Tables S5, S6, and S7.

■

RESULTS AND DISCUSSION The primary objective of the study was to establish distribution and concentration information on PFAS residues along representative ski-tracks for evaluation of regulatory measures. High mountain habitats of the Western Carpathian Mountains in Slovakia are densely covered with P. mugo (dwarf mountain pine), which usually has five different year-classes of needle generations on their branches in the Slovakian mountains and is a typical pine tree species at high-elevation locations (more than 1000 m above sea level). In Norway, P. sylvestris (Scots pine), the only native pine species, was sampled at lower elevations (max 512 m above sea level) than the Slovakian pine needle samples. In Norway, on P. sylvestris, only three yearclasses of needle generations could be collected because of the characteristic lifetime of needles in this area, which differs from P. mugo. In Norway, pine needles were collected at various ski resorts in October 2015. At many sites, the oldest generation (third year) of pine needles could not be found at the sample location.34 Therefore, a direct comparison of the age classes could only be done for the first- and second-generation needles. For this reason, second-generation needles were used for comparing all sites in this study. At selected sites, an additional comparison was done of three year-classes of needle generations for P. sylvestris and five year-classes of needle generations for P. mugo. Because the sampling of needles used different tree species and was also conducted in different countries (P. mugo in Slovakia and P. sylvestris in Norway), these potential differences must be taken into account. The plant characteristic physiology might restrict the comparability of the results provided here. However, because PFASs are reported as highly persistent, 9490


Article


Figure 4. Concentration of PFOA (with the percentage content) and total sum of PFASs (ng g−1 dw) in second-year needles in Slovakian sites (Site 1−10) and Norwegian sites (Site 11−18).

total PFASs). At Sites 4 and 9, the samples were characterized by a high abundance of PFBA (72% and 59% of total PFASs, respectively). Zilina (Site 9) is typical of an industrial town and also close to Czech and Polish borders, and thus, the influence of industrial activities (mining and manufacturing) from the Czech Republic is expected. Also, southern Poland is the most industrial part of Poland; this can influence Slovakia through atmospheric transport. However, despite high PFBA contributions, Zilina is the least-contaminated of the Slovakian sites. Krivan (Site 4) is near Zilina (Site 9), and the relative distribution of PFBA was similar. At all other Slovakian sample sites, PFOA was the predominant PFAS (>65% of total PFASs). Compared with those in the Slovakian pine needle samples, the levels from Norwegian sites had a different trend of PFAS distributions and concentrations. PFOA was the only PFAS detected and quantified in all analyzed samples (Slovakian and Norwegian), but the relative distribution differed considerably between Norway and Slovakia (Norway: 4−39%). The lowest concentrations of total PFASs were found near the capital, Oslo (Site 18). Only 3 out of 16 analyzed compounds (PFOA, PFHpS, and PFOS) were detected there. However, PFOA was found in all Norwegian sites. These results clearly demonstrate the concentration and pattern differences between Slovakian and Norwegian recreational sites. The PFOA levels found in Slovakian samples are usually 1 order of magnitude higher than those found in Norwegian samples (Figure 4). The potential for direct uptake into exposed humans can, thus, not be neglected. Nilsson et al.44 reported elevated levels in whole blood samples from professional ski wax technicians and compared their results with levels found in the general Swedish population. In professional technicians, levels were up to 112 ng mL−1 for PFOA. These occupational levels were around 45 times higher than in the general Swedish population.44 Increased concentrations of PFOA associated with skiing activities were also confirmed in other studies in blood serum, snow, soil, and air.32,48 Since 2011, the usage and production of PFOA has been banned by Norwegian regulations. National monitoring programs report stagnant levels of PFOA in a variety of environmental compartments.49 In addition, the C9−C14 PFASs are listed on Norway’s Priority List of Hazardous Substances.49 Additionally, several polyfluorinated substances are currently considered as precursors to PFOA under environmental conditions and may thus be directly transformed into PFOA during biotic and abiotic transformation in the environment. The PFOA profile observed in this study may thus be a combined result of primary emissions and transformation from precursors under environmental conditions.

(Sites 9 and 10). Zilina (Site 9) is a typical industrial location. Here, PFOA was 20% of the overall PFASs, and the absolute PFOA concentration was four times lower compared to those in the remote skiing areas. Globally, fluoropolymers manufacturing has been identified as the main direct-emission source of PFOA. These directemission sources have decreased continuously in the United States, Europe, and Japan during the past years.8 Indirect PFOA emission sources are typically the manufacturing, usage, and disposal of PFOA-containing products. This includes products containing PFOA as a direct ingredient or impurity and products in which fluorinated ingredients (precursors) may degrade to PFOA, as reported earlier.39−42 Due to the documented use of PFOA in ski waxes, their occurrence in high mountain areas may also be associated with ski resorts. Information about particular product composition and PFAS ingredients cannot be confirmed officially because the composition of ski waxes is confidential. However, investigations on exposed skiers identified high PFOA levels.43,44 The positive-reference location Kopske seddle (Site 10), where PFAS contamination was expected because of skiing activities, had a more diverse PFAS profile. PFOA at Kopske seddle was 8%, which is the lowest value from all Slovakian sites. Abundant substances from Kopske seddle and other the Slovakian samples were longer-chain PFASs (C8−C12: PFNA = 18%, PFDA = 19%, and PFDoDA = 13%). Perfluorinated carboxylic acids with a carbon chain of 11 to 14 carbon atoms are listed as substances of very high concern on the REACH candidate list because of their very persistent and very bioaccumulative properties.8 Currently, long-chain PFASs are replaced with shorter alkyl chain compounds in various PFOAcontaining products (including ski waxes). Short alkyl chain PFAS usually have similar functional properties to the longerchain compounds,45,46 including comparable environmental stability, but they are not as bioaccumulative and appear to have a less-pronounced toxicity profile.47 Kopske seddle (Site 10) is not only near the local ski resort but also close to the small town of Zdiar, which provides accommodation for tourists in many hotels. Zdiar is the gateway to the Eastern Tatra Mountains, and it is a tourist area for hiking during spring and summer. Possible sources of longer-chain PFASs may be seasonal tourist activities (impregnated outdoor clothes and products of everyday use as PFAS source). A deviation from the overall PFAS pattern was also found for Krivan (Site 4), where a relatively low level of PFOA (similar to that in Sites 9 and 10 mentioned above) was reported (26% of 9491


Article


Figure 5. Concentration of PFOA (with the percentage content) and total sum of PFASs (ng g−1 dw) in 1 to 5 year old needles in Slovakian sites (Sites 1 and 10) and in 1 to 3 year old needles at Norwegian sites (Sites 11 and 13).

All selected sampling sites are used for various skiing activities; e.g., Hafjell and Kongsberg (Sites 12 and 14) are typical Norwegian winter sports centers with cross-country ski tracks and downhill skiing facilities; Birkebeineren and Holmenkollen (Sites 13 and 15) are biathlon and cross-country arenas. Both sampling sites at Holmenkollen (Sites 13 and 14) are also known for international ski-jumping competitions and events. The sum PFASs concentration found at the positivereference site in Ås (Site 18, along the local cross-country track) was higher compared with those from the typical Norwegian ski areas, suggesting additional domestic and municipal sources in the Ås municipality. These results confirmed again that several PFAS sources exist in this type of location and that elevated levels in the environment are usually caused by a combination of product use (waterproof textiles), diffusive sources (waste and disposal sites), industrial and protective applications (i.e., the large-scale application of aqueous film-forming foams (AFFF) at firefighting facilities) and recreational activities (clothing and other impregnated textiles and surfaces, as well as ski surface preparation utilities). The negative-reference site in Oslo (Site 17) is situated in the city center near a park facility. Site 17 had low PFAS concentrations, as expected. Compared with the positive reference site in Ås, the PFAS levels were around 7 times lower (Figure 4). Trend Estimates. To assess the long-term status of the exposed environment, four sites with needles of all age classes were selected (two from each country) for comparison. After sampling, the pine needles were divided into specific year classes and analyzed separately. In Slovakia, Martinske Hole (Site 1) and Kopske seddle (Site 10) were chosen, utilizing needles of P. mugo according to five year-classes. For the Norwegian sites, Oslo winter park (Site 11) and Birkebeineren (Site 13) were selected. Pine needles from P. sylvestris were collected according to three year-classes. The summarized results are shown in Figure 5 and Table S6. The possibility of monitoring different year-classes of pine needles is considered a favorable feature for long-term temporal trend studies and environmental monitoring. In the case of pine needle samples, the uptake of airborne volatile POPs into the cellular parts of vascular plants is heavily influenced by the amount of endogenous volatile oils.51 Previous studies on pollutant levels in needles have shown that the adsorption properties for selected pollutants increase in the second year of the needle lifetime and decrease again at the end of needle lifetime because of reduced physiological fitness.22 The accumulation with age until the onset of senescence applies to all plants generally. In the case of this study, it is necessary to consider that the samples were collected

In Norway, the most-abundant PFASs were perfluorobutanoic acid (3−66%), which, however, was not detected at Hafjell (Site 12) and Oslo (Site 17); and perfluoropentanoic acid (PFPA, 21−85%), which was not detected in and near Oslo (Sites 15−17) as well as Kongsberg (Site 14). The highest concentrations of PFPA were found in Hafjell (Site 12) and Ås (Site 18), where PFBA was absent or is low. In general, shorterchain PFASs (chain length C4−C7) are more common in Norway. PFNA, PFDA, PFUnDA, and PFTeDA were not detected, which obviously reflects again the effect of the Norwegian national PFAS regulations.49 Distinct differences in distributions between the sampling sites in Norway and Slovakia were also found for perfluoroheptanesulfonate (PFHpS). In Norway, PFHpS was found at all sampling sites except Kongsberg (Site 14), whereas in Slovakia, PFHpS was detected only at Kopske seddle (Site 10). Comparison of Overall Concentrations. The comparison of the total PFAS concentrations at Slovakian and Norwegian sites (Figure 4 and Table S5) revealed higher sum PFAS levels in Slovakia compared to those in Norway. The highest PFAS levels in all samples were found at Krivan (Site 4). However, the dominance of PFBA (72% of total PFASs) most probably reflects direct atmospheric transport from Polish industrial areas near rural Krivan. Contamination at other Slovakian sites is similar with respect to levels and distribution patterns. PFOA dominated the remaining Slovakian Sites 1−3 and 5−8. At first glance, the reference sites showed obvious differences in patterns. However, the relatively high concentration in Zilina (Site 9, negative-reference site) was most probably caused solely by a high PFBA contribution (59%). Except for PFBA concentration, Zilina, as an urban site, was obviously the least-contaminated Slovakian site. For the Kopske seddle Mountain station (Site 10, positive-reference site), high levels for all perfluorinated compounds (C6−C10) were confirmed. For Norwegian sites, the highest levels were reported for Oslo winter park (Site 11) close to Voksenkollen. The pine trees selected for sampling were usually located in the vicinity of the ski school, near ski tracks but as far as possible from the city center, and industrial activities as well as residential areas. Ski schools are usually also locations for technical support for skiers and places for ski rentals. Therefore, ski waxes may be more widely used in these locations. Our results confirmed also a recent study by the Norwegian Institute for Air Research (NILU) in which elevated levels of PFASs were identified in earthworms from the Voksenkollen area, located in the vicinity of our sampling site.50 The presence of PFAS in this location is exclusively explained by the comprehensive usage of ski waxes due to skiing activities.50 The contamination levels for the other Norwegian sites are low and found to be similar for all samples. 9492


Article

Environmental Science & Technology in late spring in Slovakia and in autumn in Norway. A large seasonal difference in the accumulation rate during spring and summer can be observed for selected POP compounds.51 Kylin et al.52 proved that the accumulation of α-HCH in P. mugo needles from two different sites had a general trend of increasing concentrations during the needle lifetime, except for the last available year, when senescence had started. The theory that the previous year-class starts losing contaminants52 was confirmed for P. mugo (Sites 1 and 10) at the Slovakian sites. This assumption was also confirmed for selected POP compounds in a recent publication of a 4 year long study with a considerably higher number of samples.53 However, considerable variations in the PFAS distribution profiles limit the overall validity of this hypothesis. Thus, these results should only be considered as indicative. In addition, immediate changes in environmental conditions (i.e., temperature variations, irradiation etc.) may cause increased or decreased sorption capability or even desorption for selected pollutants and thus obscure the age-related adsorption profiles (increased concentration variation). Because sampling was carried out in the late spring and early summer period, another explanation for the concentration variation can be also due to various terpenoids content that are present in the needles during the warm and dry summer months.51 For P. sylvestris (Sites 11 and 13), Kylin et al.27 reported that the adsorptive capability for selected persistent pollutants increases during the entire lifetime of needles. This assumption is supported by these results. A considerable increase in concentration of sum PFASs is reported for the second- and also for third -year needles (P. sylvestris) for both Norwegian sites. Despite the indicative results, this study is only an evaluation of two sites from each sampling country, and therefore, the number of samples is small, and the results serve only as a semiquantitative assessment of trend estimates. Altitudinal Patterns and Distance to Primary Sources. For selected Slovakian sites (Sites 1, 2, 3, and 4), the samples were collected along an elevation gradient, which also reflects distance to potential primary sources. Sites 1 and 2 as well as 3 and 4 were compared according to the previously described site characteristics. The results were examined with respect to prevailing wind direction and distance to potential contamination sources, considering the contribution of long-range atmospheric transport (from industrial, recreational, and domestic sources). For Sites 6 and 7, sample transects followed an elevation gradient (Figures 6 and 7 and Table S7). For these sites, the lowest point was sampled at 1515 m above sea level and the highest at 1756 m above sea level. For a more-complete source characterization at the area, the PFAS levels were combined with previously reported concentrations of legacy persistent organic pollutants (POPs: polychlorinated biphenyls, PCBs; hexachlorocyclohexane isomers, HCHs; dichlorodiphenyltrichloroethane derivatives, DDTs) from the same location22 (Table S7). The expectation was that ski preparation (i.e., ski waxes) and outdoor textile products are the predominant source of PFAS in the locations used for winter sport activities, whereas POP contamination is usually caused by diffusive emissions from municipal installations and long-range transport. Therefore, we expected different trend patterns for each pollutant group. This was expected to provide complementary information for complete source apportionment of the identified pollution profiles. The results of a previously

Figure 6. Concentration of PFOA and total sum of PFASs (ng g−1 dw) in 2 year old needles in the Slovakian sites Martinske Hole Mountain (Sites 1 and 2), Krivan Mountain (Sites 3 and 4), and Sivy vrch Mountain (Sites 6 and 7), depending on the slope of exposure and elevation.

Figure 7. Concentration of total sum of PCBs, HCHs, and DDTs (pg g−1 dw) in 2 year old needles in the Slovakian sites Martinske Hole Mountain (Sites 1 and 2), Krivan Mountain (Sites 3 and 4), and Sivy vrch Mountain (Sites 6 and 7), depending on the slope of exposure and elevation.

reported study22 confirmed the highest concentrations of POPs at areas in the vicinity of industrial locations. Long-range transport was identified as the second most important source in the high mountain habitats. The highest concentrations were found in areas with prevailing winds from the Czech Republic and Poland, as these are considered potential source areas. However, no clear influence of long-range atmospheric transport was observed for PFAS at the selected skiing resorts. This may indicate the dominance of local pollution sources. For Martinske Hole Mountain (Sites 1 and 2), similar trends were found for PFASs and POPs (Figures 6 and 7). The sampling site was located on the eastern side of the mountain and is highly influenced by contamination transported from residential and industrial areas around the city of Martin. A large popular skiing resort is situated close to the eastern mountain site in the vicinity of the city. This may be the reason for the elevated POP and PFAS concentrations (caused by potential local contamination sources). Low PFAS levels were observed at the sampling location on the west side of the mountain (given the absence of a local ski resort). Here, longrange atmospheric transport is considered the only source of PFAS contamination. The concentration of POPs found in the sampling regions was slightly elevated. Thus, a considerable contribution from long-range atmospheric transport from Czech Republic may be assumed. 9493


Article


This study provided scientifically sound information on the comprehensive value of pine needles as passive samplers and as a monitoring tool for spatial and temporal PFAS distribution profiles, as well as those of other priority organic pollutants. We therefore encourage regulatory authorities and the research community to further develop the concept and consider this analytical strategy for implementation in long-term national pollutant monitoring programs. Specifically, P. mugo as sample medium has previously been used only once, by our research group.22 This study provided sampling and analytical approaches suitable for monitoring in response to national and international regulations and legislative restriction for the application of chemicals, as demonstrated here for PFASs.

The opposite trend was observed at the Krivan Mountain range (Sites 3 and 4). The concentration of POPs at these sites showed no marked differences between the northern and southwestern locations (Figure 7), suggesting that industrial activities are contributing to the POP contamination at both locations in a similar way. However, for PFAS patterns, several distinct differences in source and distribution profiles were identified. On the north side of the mountain, a large ski resort is present. Therefore, we assume that this could be a major reason for the elevated PFAS levels found in the needles. The PFAS concentration, however, is increased slightly (compared to the Martinske Hole site) toward the southwestern sampling locations (Figure 6), and the increasing PFAS patterns are mainly dominated by PFBA and PFPA, indicating primary sources. At Sivy vrch Mountain (Sites 6 and 7), a transect of samples along an elevation gradient was collected in 2014 and composed of five sites (1515−1756 m above sea level).22 Sivy vrch Mountain is influenced by air masses moving from the north and the southern side, thereby passing industrial locations as well as densely populated areas as potential source areas. Thus, we assumed that the air-mass movement and direction from the north side was contaminated by the northern industrial area, while air masses on the south side are in contact with emissions from the residential area. Air masses from the northern side pass over the mountain ridge and are further contaminated by emissions from the southern side. These air masses then turn in the opposite direction during favorable meteorological conditions and move back again to the lowest transect site.22 Consequently, the highest POP contamination was found at the lowest point of the transect. The second-highest concentrations were found on the highest point of transect (Figure 7). The lowest and highest point of transect, again, were chosen for PFASs analysis. In contrast to the POP profiles (Figure 7), the PFAS levels do not show marked differences between the lowest and highest sampling points (Figure 6). The PFAS levels reported are higher at the highest transect point. Therefore, the industrial areas in vicinity of Sivy vrch Mountain do not affect the concentration of PFAS along the mountain range. Because only four sites (Sites 1, 2, 3, and 4) were used to evaluate the influence of distance to primary sources and only two sites (Sites 6 and 7) for evaluation of transects, these results are not strictly conclusive. These results provide a basic view on the influence of local sources, long-range transport, slope of exposure, and elevation on concentrations of selected POPs. Confirmation of these hypotheses requires moredetailed sampling and research on PFAS accumulation in pine needles, mainly because this is the first study concerning the presence of PFAS compounds in pine needles. In conclusion, significant regional and local differences in PFASs distribution patterns and concentrations between ski resorts in Slovakia and Norway were identified. The nationspecific PFAS regulation contributed specifically to the PFAS profiles identified in the respective countries. This study provided important information on PFAS source characterization in remote ski areas. Source elucidation were performed for a characteristic set of priority pollutants, including legacy POPs and PFASs. This evaluation further confirmed the contribution of multiple primary sources to the overall contamination profile of the respective recreational skiing resorts.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b02264. Tables showing the geographical coordinates and elevation of sampling sites in Slovakia and Norway; MS parameters for PFASs compounds; percentage recovery of native PFASs; concentration of PFASs in ng g−1 dw and

Pine Needles for the Screening of Perfluorinated Alkylated

Recommend Documents