PFAS in recipient sediment of a military airport

Örebro University, Sweden

PFAS in recipient

sediment of a military

airport

Lovisa Johansson Blomér loviisa_blomeert@hotmail.com Supervisor; Ingrid Ericson Jogsten Examiner; Anna Kärrman

1

Table of content

2.1 Per- and polyfluoroalkyl substances 5-6

2.2 Application of PFASs 6-7 2.3 Production 7 2.3.1 Electrochemical fluorination 7 2.3.1 Telomerization 7-8 2.4 Regulations 8 2.5 Objective 8

3.0 Materials and method 9

3.1 Samples 9

3.2 Chemicals 10

3.3 Method validation 10

3.4 Sample extraction and clean up 10

3.5 Instrumental analysis 10

3.6 Quality control and quality assurance 11

4.0 Results and discussion 11

4.1 PFASs concentration in Vissbäcken and outlet 4 (utlopp 4) 11-14

4.2 PFOS concentration in sediment core 15-19

4.3 PFAS association with metal ions 20-22

5.0 Conclusion 23

6.0 Acknowledgment 24

References 25-26

Appendix A 27-31

2

List of Figures

Figure 1. Sampling sites from Lake Vänern, Vissbäcken and Outlet 4 (utlopp 4) 9

Figure 2. Homologue patterns for detected PFASs in Vissbäcken and outlet 4 12

Figure 3. Detected PFASs concentrations in Vissbäcken 13

Figure 4. PFOA, PFOSA and PFOS concentrations in Vissbäcken 13

Figure 5. PFAS concentration in outlet 4 (utlopp 4) 14

Figure 6. PFOA, PFOSA and PFOS concentrations in outlet 4 (utlopp 4). 15

Figure 7. PFOS concentration from S-SED1:1 17

Figure 8. PFOS concentration from S-SED2:1 17

Figure 9. PFOS concentration from S-SED3:1 18

Figure 10. Ranging levels of PFOS in sediment cores showing different sampling sites, depth and values <LOD. PFOS concentration is calculated in wet weight 18

Figure 11. Association between PFOS and metal ions in sediment core S-SED1:1. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units 21

Figure 12. Association between PFOS and metal ions in sediment core S-SED2:1. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units 21

Figure 13. Association between PFOS and metal ions in sediment core S-SED3:1. Here, the metals correlates, but not PFOS. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units 22

3

Abstract

Per- and polyfluoroalkyl substances, PFASs, are highly fluorinated substances where the carbon chain is fully- or partly fluorinated. A functional group is coupled at the end of the carbon chain which gives PFASs their different properties. PFASs have been used in aqueous film forming foams (AFFFs) to decrease the surface tension of water and form a film on the fuel surfaces. AFFF is one of the main sources of PFASs pollution in the environment. A previous study has shown high PFASs concentrations in surface water in Lake Vänern. This study has analysed PFASs in sediment samples. The main detected PFASs was perfluorooctane sulfonate, PFOS, with concentrations below limit of detection to 51700 pg/g wet weight (ww). The compound detected in the highest concentration in Vissbäcken was PFOS at 7290 pg/g ww, this was followed by 6:2 fluorotelomersulfonate , 6:2FTS, at 516 pg/g ww. In outlet 4 (utlopp 4), PFOS had the highest concentration at 51800 pg/g ww, followed by perfluorohexane sulfonate, PFHxS, at 1790 pg/g ww. The only detected compound in the sediment cores was PFOS with approximately 100 pg/g ww. The high concentration of PFOS might be due to extensive use with subsequent release of firefighting foam in the area and degradation of other PFAS substances into PFOS.

4

Sammanfattning

Per-och polyfluoroalkylsubstanser, PFASs, är en grupp högfluorerade ämnen där kolkedjan är helt eller delvis fluorerad. En funktionell grupp kopplas i slutet av kolkedjan vilket ger PFASs dess olika egenskaper. PFAS har använts i vattenhaltiga filmbildande skum (AFFF) för att minska ytspänningen av vatten och bilda en film på bränsleytan. AFFF är en av

huvudkällorna för PFAS-föroreningar i miljön. En tidigare studie har visat på höga koncentrationer av PFASs i ytvatten i Vänern. I denna studie har PFASs analyserats i

sedimentprover. Den huvudsakliga detekterade PFAS var perfluoroktansulfonat, PFOS, med koncentrationer under detektionsgränsen och till 51700 pg /g våtvikt. Den högst detekterade föreningen i Vissbäcken var PFOS med 7290 pg/g våtvikt, detta följdes av 6:2

fluorotelomersulfonat, 6:2 FTS, med 516 pg/g våtvikt. Den högst detekterade föreningen i utlopp 4 var PFOS med 51800 pg/g våtvikt, följt av perfluorhexansulfonat, PFHxS, med 1790 pg /g våtvikt. Den enda detekterade föreningen i sedimentpropparna var PFOS med ungefär 100 pg/g våtvikt. Förekomsten av PFOS i sediment i sjön Vänern kan bero på omfattande användning med efterföljande utsläpp av brandbekämpningsskum i området och nedbrytning av andra PFAS-substanser till PFOS.

5

1.0 Background

A previous study has shown high per- and polyfluoroalkyl substance, PFAS, concentrations in surface water in Lake Vänern (Försvarsmakten, 2015). In 2015, a national work was

performed to locate PFASs in soil and water at airports and old fire drill locations. Analysis of PFASs at Skaraborgs Air command F7 has been done by the mandate from the armed forces (Försvarsmakten). The study showed high concentrations of PFAS in soil and groundwater at the sites were the fire drills had taken place. It also showed environmental distribution of the substance. However, the pollution was mainly distributed through the surface water and not groundwater. The study gave an overview of the situation. At present, there is still no information on PFAS concentrations in the sediment (Lidköping kommun, 2017) Vänern is Sweden’s largest lake with an area of 5 650 km2 and length at 4 800 km (lakevanern.se)

A potential source of PFASs in the environment are fire drill locations where firefighting foam containing PFAS are used (Ahrens et al., 2015).

2.0 Introduction

2.1 Per- and polyfluoroalkyl substances

Per- and polyfluoroalkyl substances, also known as PFASs, is the generic term for highly fluorinated substances where the carbon chain is fully- (perfluorinated) or partly fluorinated (polyfluorinated) (Buck et al., 2011). A functional group is usually coupled at the end of the carbon chain which gives PFASs their different properties. The fluorinated tail is hydrophobic (water repellent) and the functional group is hydrophilic (water soluble). Due to the strong bond between the carbon and flourine atom, PFASs will be chemically and thermally stable (Kissa, 2001).

The partly fluorinated surfactants have different properties over the perfluorinated surfactants. Due to the hydrocarbon segment, the polyfluorinated surfactant is more soluble in more commonly used solvent, has a lower melting point, have reduced volatility and decreased acid strength of fluorinated acids (Kissa, 2001).

The compounds carbon chain length will affect its environmental behaviour. For example shorter chains (<C8) have been found more in lettuce leaves, while longer chains (≥C8) has been found more in the roots (Banzhaf et al., 2016). Further, studies of shorter chain

molecules show that perfluorobutane sulfonate, PFBS, has a faster elimination time and is less toxic and bioaccumulative than longer chain compounds that have higher toxicity due to the accumulation in the liver (Stahl et al., 2011).

6

PFASs are highly stable under natural condition and some precursor PFASs can degrade but form highly stable PFAS instead, such as for example perfluorinated carboxylic acids (PFCAs) and perfluoroalkane sulfonic acids (PFSAs). These are collectively termed perfluorinated alkyl acids (PFAAs). PFAAs can be transported long distances in water and aerosol due to their water solubility. The legacy and ongoing use of PFASs has led to a global distribution of PFAAs in the environment. The high persistence of PFAAs causes the

accumulation of PFAAs in the environment, for example in groundwater and sediment. (Wang et al., 2017).

Several studies have shown levels of PFAS in food, drinking water, surface water, sediment, flora, fauna and humans (Eriksson et al., 2013; Banshaf et al., 2016; Kärrman et al., 2004; Hu et al., 2016). PFASs have been found in sediment at several places such as Pearl river

Estuary, South China, Lake Chaohu, Eastern China, several lakes and rivers in France and from Gulf of Gdańsk, the Baltic Sea (Gao et al., 2015; Qi et al., 2015; Munoz et al., 2015; Falandysz et al., 2012). PFASs in sediment can be influenced by several factors such as the carbon length, presence of metal ions, particle size and total organic carbon. The study from Pearl river Estuary, South China, showed a spatial distribution between PFAS and metal ions, this could indicate a similar source of the pollutants (Qi et al., 2015).

It has been shown that cohesive forces between particles in the sediment leads to formation of aggregates of adequate size and density for deposition. In high energy environments, example in low water depths relative to large fetch, the sediment is not deposited. The water’s quality, movement and distribution will influence the deposition, as well as the area’s appearance, the deposition will also increase with increasing water depth. Mercury is an indicator of the accumulation process, the distribution of mercury in water is coupled with the water´s flow pattern. The mercury has an affinity to the small organic particles and aggregates of particles which will deposit with them (Håkansson & Larsson, 1976).

2.2 Application of PFASs

Since the 1950s, PFASs has been used in many products due to their water and fat repellent abilities (Naturvårdsverket, 2016). Almost one fifth of the substances have surface active properties (Kemi, 2015). PFAS is also found in products such as camping equipment, non-stick cookware and firefighting foam (Naturvårdsverket, 2016). PFASs have also shown to accumulate in the food chain, mostly in protein-rich tissue, and in the environment (Bossi et al., 2014).

The aqueous film forming foams (AFFFs) uses synthetic chemicals instead of protein-based materials. The surfactants will decrease the surface tension of water and form a film on the fuel surface (Kissa, 2001)

PFASs in AFFF has been produces since the late 60s and used since the 60s and early 70s (Anderson et al., 2016; Ahrens et al., 2015). One problem with AFFF is that it is usually used

7

in liquid form during a short period of time, which can increase the risk of pollution of PFASs in aqueous environments. In one study analysing PFASs in water and sediment in surrounding areas of Stockholm Arlanda airport, where a fire training facility is located north of the

airport, perfluoroocatane sulfonate (PFOS), perfluorohexane sulfonate (PFHxS) and perfluorooctanoic acid (PFOA) were dominating in the water sample from Lake Halmsjön. PFOS were even more dominating in the sediment. This is an indication of compound-specific distribution of PFASs in multi compartment environments (Ahrens et al., 2015).

2.3 Production

Because fluorine is extremely reactive, direct fluorination is not commonly used for

commercial synthesis. The energy needed to form the C-F bond, which is one of the strongest bonds, is about 460 kJ/mol which exceeds the C-C bond at 348 kJ/mol, the strength of the carbon bond is also what makes the molecule so persistent in the environment. When synthesising fluorinated surfactants it is important to control the reaction because fragmentation of the substrate can take place. Different production ways to fluorinate

surfactants is used today, among them electrochemical fluorination and telomerisation (Kissa, 2001).

2.3.1 Electrochemical fluorination

Electrochemical fluorination is one process for synthesising PFASs. It uses liquid hydrogen fluorine, HF, to dissolve the organic substrate to be fluorinated. An electric current is passed through the hydrogen fluorine at a voltage between 5-7 V. All the hydrogen atoms in the molecule are replaced with fluorine at the cathode, but the functional groups is retained (Kissa, 2001). The major product produced by ECF is perfluorooctane sulfonyl fluoride (POSF). EFC is a crude process which leading to variety in the product with approximately 70% straight chain POSF and 30 % branched isomers (Buck et al., 2011).

2.3.2 Telomerization

Telomerization is another process for synthesizing PFASs where perfluoroalkyl iodine is used to react with tetrafluoroethylene to yield a mixture of perfluoroalkyl iodides with longer perfluorinated chains. The starting iodide is called telogen and the tetrafluoroethylene is called taxogen. The product perfluoroalkyl iodide is then further reacted in a second step. The product from the second reaction is then further reacted to create fluorinated substances, example fluorotelomer alcohols (FTOHs). When employing a linear telogen and taxogen, a linear product will form. When employing a telogen or taxogen that are branched or have an odd number of carbon atoms, the product will be branched or contain an odd number of carbons (Buck et al., 2011).

8

2.4 Regulations

The Stockholm Convention on Persistent Organic Pollutants, POPs, was adapted in 2001 and enforced in May 2004. Due to the harmfulness on humans, wildlife and the environment, the main purpose of the Stockholm Convention is to eliminate or reduce POPs release and distribution in the environment (Stockholm Convention, 2008). PFOS and PFOS-related substances were added to the Stockholm Convention in 2009 (Stockholm Convention, 2008). Since June 27 2008, PFOS has been forbidden, with some exceptions, in the European Union, EU, due to its persistent, bioaccumulative and toxic properties. PFOA is still not banned in the EU (Directive 2006/112/ECOF).

The regulations have reduced the concentrations of some PFASs, mainly PFOS, in the

environment. But aqueous film forming foams (AFFF), which contains PFASs, is still used by the military and civilian airfields. The ongoing use of AFFF and other products containing PFASs, will continue the spread of PFASs in the environment (Banzhaf et al,. 2016).

2.5 Objective

The objective of this project is to examine the presence of PFASs in sediment in Lake Vänern outside the known point source Skaraborgs military airport. Previous analysis of water

samples have shown high PFAS concentrations. Also included in this work is to study the distribution at various locations and depths.

Another student project will study the distribution of metals in the same samples, making association analysis and multivariate statistical analysis possible. This could provide relevant information regards environmental behaviour of PFASs and possibilities of analytical

determination depending on the simultaneous presence of metals. The questions of issue were:

1. What PFASs are found in the sediment in Lake Vänern and what are the concentrations? 2. Do concentration vary with location of sampling point and depth of sample?

9

3.0 Materials and method

To avoid contamination, all equipment was pre-washed with methanol. Sample preparation, extraction and clean up was performed in fume hoods at MTM laboratories at Örebro University. The samples were stored in refrigerators at -5°C while sample extract where stored in -18°C after sample clean up. The samples were analyzed one to five weeks after sampling.

3.1 Samples

The samples were collected by NIRAS on 2017-03-28. Three sediment cores (SED1. S-SED2, S-SED3) with 34 cm depth were collected and sliced; 2 cm for the first 10 cm and 4 cm for the remaining 20 cm. The last 4 cm was removed to avoid possible contamination. One outlet (Utlopp 4), and a sample at Vissbäcken were collected on 2017-03-29, these samples were collected using a grab sampler and did not contain any sediment cores from different depths. The core sediment was sandy between 0-4 cm and muddy from 4-30 cm. The grab samples were more organic material. For pictures from the sampling, see Appendix A. Figure 1 shows sampling sites.

10

3.2 Chemicals

HPLC methanol grade and LC-MS methanol grade was bought from Fisher Scientific (Loughborough, UK). The ENVI-Carb SPE Bulk, N-Methylpiperidine (1-MP) and

ammonium acetate (NH4Ac) was bought from Sigma-Aldrich (St. Louis, USA). Solid sodium

hydroxide was bought from KEBO Lab. Hydrochloric acid, 0.1 M, was bought from Scharlab (Barcelona, Spain). The glacial acetic acid was bought from Merck KGaA (Darmstadt, Germany).

3.3 Method validation

The method was tested by analysing three replicates of samples from this study and one reference sample from a previous study at MTM. The results are shown in table 12 in Appendix B.

3.4 Sample extraction and clean up

One gram of the sediment sample was weighed in PP tubes and 10 µl of 0.2 ppm internal standards were added. Sodium hydroxide, 2 ml 0.2 Min methanol was added and the sample was mixed using a vortex mixer. Next, 4 ml methanol was added. The samples were then ultra-sonicated for 15 minutes and shaked for 15 minutes. Hydrochloric acid, 400 µl of 1 M, where used to neutralize the samples. The samples were centrifuged for 10 minutes at 6000 rpm. The supernatant was transferred to PP tubes containing 50 mg Envi-carb and 100 µl glacial acetic acid. The samples were re-extracted by adding 4 ml methanol, vortexed, shaked for 15 min and then centrifuged for 10 min at 6000 rpm. The supernatant was transferred to the PP tubes. The samples were evaporated to under 1 ml and filtrated into LC vials with a 0.2 µm filter. The LC vials had been spiked with 10 µl 0.2 ppm recovery standard. The sample extracts were evaporated to 200 µl and 300 µl mobile phase was added before analysis. For every 10 samples, one blank and one reference sample was used.

3.5 Instrumental analysis

Analysis were performed with a Waters Acquity UPLC-Xevo TQ-S tandem mass spectrometer with electro spray ionisation in negative ion mode. The LC column was a 100 mm BEH C18 with 2.1 mm in column diameter and 1.7 µm particle size. The column temperature was set to 50 °C, the desolvation temperature at 350 °C, capillary voltage at 3.0 kV, the desolvation gas flow at 800 L/Hr and cone gas flow at 150 L/Hr. A total of 10 µl of sample was injected.

For the separation of PFASs, mobile phase A consisted of 70 % H2O, 30% MeOH, 2 mM

NH4Ac and 5 mM 1-MP. Mobile phase B consisted of 100% MeOH, 2 mM NH4Ac and

11

phase A and was then ramped up to 100% mobile phase B over a total runtime of 18 minutes. The samples were processed in Masslynx.

3.6 Quality control and quality assurance

For calibration, a seven-point calibration curve was used with 0.02, 0.1, 0.2, 2, 10, 20 and 40 ng/ml. A standard used for single point calibration was prepared with each set of samples extracted. This one point calibration standard was then compared to the seven point

calibration curve. Native standard was used for recovery experiments. An internal standard and a recovery standard was used for quantification. One blank and one reference sample was used for every 10 samples.

The target product ions (2-3 per PFAS, if present) were compared and was within a 25% difference to ensure the quality of the quantification. The limit of detection, LOD, was determined using three times the average blank concentration of detected peaks or background noise. The LOD values can be seen in Table 2 in Appendix B.

4.0 Results and discussion

Out of 24 analysed PFASs, a maximum of 21 and a minimum of one PFAS was detected in each sample. PFOS was the dominant PFAS in all sediment samples. Perfluorobutanoic acid (PFBA), perfluorononanoic acid (PFNA) and perfluorooctadecanoic acid (PFOcDA) was not detected in any sediment sample due to low recovery and within 25 % for positive

quantification. L-PFOS was the analysed PFOS.

The most abundant product ions for quantification and confirmation were chosen for each target analyte. Positive quantification criteria was set to quantification ion and confirmation ion being within 25 %.

In another student project the water content of the samples was determined (Table 3, Appendix B).

4.1 PFASs concentration in Vissbäcken and outlet 4 (utlopp 4)

Of the 24 PFASs that were analysed, 16 PFASs were detected in Vissbäcken. The PFOS concentration, as seen in Figure 2 and 4, was dominant in Vissbäcken. This was followed by 6:2 fluorotelomersulfonate, 6:2FTS, (516 pg/g ww) and PFOSA (416 pg/g ww). The

compound having the smallest contribution to total PFAS was the four carbon chain length perfluorobutane sulfonate , PFBuS, contributing with only 0.17 % and a concentration of 16,4 pg/g ww of the total PFAS concentration. The longest detected compound was

perfluorododecane sulfonate, PFDoDS, (177 pg/g ww). The presence of both long chain and short chain sulfonates might indicate that the source of these contaminants is from the airport. PFOS has been a main component in AFFF and during electrochemical fluorination both

12

isomers and other types of sulfonates can be formed. The ratio between the highest detected compound (PFOS) and second highest detected compound (6:2FTSA) was 7%.

Long chain carboxylates as PFDA and PFDoDA were below the LOD (table 2 Appendix B) and could not be quantified. PFHxA was also below LOD and could not be detected. Of the 24 PFASs that were analysed, 21 PFASs were detected in outlet 4 (utlopp 4). The composition pattern of PFASs differed only slightly between the sampling points of outlet 4 and Vissbäcken. As seen in Figure 5 and 7, the PFOS concentration in outlet 4 (utlopp 4) was dominant with 85 % and51800 pg/g ww of total PFASs concentration, followed by PFHxS (1790 pg/g ww) and PFOSA (1730 pg/g). The compound having the smallest contribution to total PFAS was PFBuS contributing with only 0.13 % and a concentration at 80.5 pg/g ww of the total PFAS concentration. PFBuS was also the shortest compound detected and

perfluorohexadecanoic acid, PFHxDA, (44.9 pg/g ww) the longest. The ratio between the highest detected compound (PFOS) and second highest detected compound (PFHxS) was 3.4 %. This corresponds with other studies where PFOS also is the predominant PFAS found in sediment with 86% of total PFASs (Ahrens et al., 2015).

One of the main sources of PFASs pollution in the environment is fire-fighting foam from fire drill locations, such as airports (Ahrens et al., 2015). This is probably the main reason for high PFASs concentrations in Vissbäcken and outlet 4 (utlopp 4). This could also be the reason for the large composition variation. The dominance of PFOS in both Vissbäcken and outlet 4 (utlopp 4) is probably due to AFFF with PFOS as a component. It could also be due to break down of other PFASs into PFOS.

In another student project, PFASs concentrations have been detected in surface water (Table 4, Appendix B). PFOS was the most abundant PFASs with concentrations of 4160 ng/l in Vissbäcken and 114 ng/l in outlet 4. This could be an indication of compound specific distribution for PFOS (Ahrens et al., 2009)

13

Figure 2. Homologue patterns for detected PFASs in Vissbäcken and outlet 4.

Figure 3. Detected PFASs concentrations in Vissbäcken, PFOS, PFOA and PFOSA are included in another figure due to high concentrations of PFOS.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Vissbäcken Outlet 4

Vissbäcken and outlet 4

PFPeA PFBuS PFHpA PFPeS PFHxA PFHxS PFHpS PFDA PFUnDA PFNS PFOA PFOSA PFOS PFDS PFDoDA PFTrDA PFDoDS PFTDA PFHxDA 6:2FTSA 8:2FTSA

0 100 200 300 400 500 600 Co nce nt ra tio n p g/g w w

Vissbäcken

PEPeA PFBuS PFHpA PEPeS PFHxS PFHpS PFUnDA

14 Figure 4. PFOA, PFOSA and PFOS concentrations in Vissbäcken.

Figure 5. PFAS concentrations (pg/g ww) in outlet 4 (utlopp 4), PFOS, PFOA and PFOSA are included in another figure due to high concentrations of PFOS.

0 1000 2000 3000 4000 5000 6000 7000 8000 Co nce nt ra tio n p g/g w w

Vissbäcken

PFOA PFOSA PFOS

0 200 400 600 800 1000 1200 1400 1600 1800 2000 Co nce nt ra tio n p g/g w w

Outlet 4 (utlopp 4)

PFPeA PFBuS PFHpA PFPeS PFHxA PFHxS PFHpS PFDA PFUnDA PFNS PFDS PFDoDA PFTrDA PFDoDS PFTDA PFHxDA 6:2FTS 8:2FTS

15 Figure. 6 PFOA, PFOSA and PFOS concentrations in outlet 4 (utlopp 4). Note that the PFOA concentration is at 312 pg/g ww.

4.2 PFOS concentrations in sediment cores

The only PFAS substance that could be detected in sediment core samples from Lake Vänern was PFOS. The PFOS pattern differs from different sites and depths (See Figure 11). The PFOS concentration varied between concentrations under LOD and 323 pg/g ww. The limit of detection, LOD, was 81 pg/g as seen in Table 2 in Appendix 2. The remaining PFASs were not detected due to concentrations <LOD, poor recovery or more than 25% difference in target ions.

The PFOS concentration in S-SED1:1 (figure 8) had the highest concentration (144 pg/g ww) at a depth between 14-18 cm. The ratio between the highest detected concentration (14-18 cm) and second highest detected concentration (0-2 cm) was 79 %. Samples from three depths (0-2 cm, 2-4 cm and 14-18) were over LOD. With concentrations at 114 pg/g ww, 101 pg/g ww and 144 pg/g ww, respectively.

The PFOS concentration in S-SED2:1 (figure 9) had the highest concentration at a depth at 0-2 cm with 0-218 pg/g ww. The ratio between the highest detected concentration (0-0-2 cm) and second highest concentration (8-10 or 10-14) was 79 %. Depth 8-10 and 10-14 had the same detected concentration (173 pg/g ww). Only samples from two depths had a concentration under LOD (22-26 cm and 26-30 cm).

The PFOS concentration decreased between depth 0-6, after which it increased between depth 6-14 cm, and then decreased between 14-30 cm.

The PFOS concentration in S-SED3:1 (Figure 10) had its highest concentration at depth 6-8 cm (323 pg/g ww), which also was a peak in PFOS concentration. The ratio between the highest concentration (6-8 cm) and second highest (2-4 cm) was 41 %. The concentrations

0 10000 20000 30000 40000 50000 60000 Co nce nt ra tio n p g/g w w Substance

Outlet 4 (utlopp 4)

16

between 8-22 cm were <LOD. The concentration between 0-6 cm of depth was similar with concentrations at 104 pg/g ww, 133 pg/g ww and 98.5 pg/g ww respectively.

There is a difference in PFOS concentration between sampling sites and sediment depth. However, the pattern is similar between S-SED1:1 and S-SED2:1. This indicates the same source and environmental distribution. Exact usage of AFFF is not available information else the profiles might reflect different usage over time. The reason for that S-SED3:1 is different can be due to different distribution or source. PFOS has one of its highest concentration in the surface sediment, and will decrease in deeper depth. The concentration will then increase some. The PFOS concentration was then undetected at 22-30 cm of depth.

S-SED1 has a general lower PFOS concentration than S-SED2 and S-SED3, this might be due to the main emission reaching the area origins from outlet 4, which has passed a cleaning facility to try to lower the PFOS amount.

The levels of PFOS varies over time as indicated by different concentrations at different sediment depths. This can be due to difference in use of firefighting foam over time, the use of different contents in the firefighting foam, as well as the ability of the treatment plant to handle PFAS contaminants.

Earlier studies from the area have shown that the environmental distribution of PFASs from the military airport area was mainly in surface water and not in ground water (lidkoping.se). When the substances will spread through the surface water and reaches Vänern the dilution effect will be high. This might be the reason for the low detection rate of PFASs in the sediment cores. The locations of sampling points might also reflect this with SED 1 and 2 in the more enclosed bay of Brandsfjorden while SED 3 is located more towards the open part of this large lake.

Figure 7. PFOS concentration from S-SED1:1. 0 20 40 60 80 100 120 140 160 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 Co nce nt ra tio n g/ g w w Depth, cm

17 Figure 8. PFOS concentration S-SED2:1.

Figure 9. PFOS concentration S-SED3:1. 0 50 100 150 200 250 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 Co nce nt ra tio n p g/gw w Depth, cm

PFOS concentration S-SED2:1

0 50 100 150 200 250 300 350 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 Co nce nt ra tio n p g/g w w Depth

18 Figure 10. Ranging concentrations of PFOS in sediment cores showing different sampling sites, depths and values <LOD. PFOS concentration is calculated in wet weight.

Figure 10 shows the differences in PFOS concentration between sites and values <LOD. PFOS concentration for dry weight samples is seen in figure 15, Appendix B. The water content in the samples was between 28-57%. When calculating the PFOS concentration to dry weight, more samples could exceed or be close to the LOD value at 81 pg, this because only PFOS was calculated on dry sediment and the water was removed. From the wet weight samples 13 were <81, the PFOS concentration at dry weight had 5 samples <81, and 2 samples that were close to the LOD value.

The PFOS concentration differs from other studies. In the study from Pearl river estuary in South China PFBS and PFHxS was the dominant PFASs found in the sediment which could be explained by the increasing use of PFBS and PFHxS as a substitute for PFOS. The PFOS concentration varied from <LOD to 320 pg/g dw (Gao, et al., 2015). In another study from Pearl river delta region in south china, PFOS was the dominant PFASs found with a concentration at 11.4 ng/g dw, followed by PFOA (Chang-Gui et al., 2014).

In another study from the Baltic Sea, PFOS was the main PFASs found in core sediment. However, PFHxS and PFOA were also detected. Also, PFDoDA, PFUnDA, PFDA and PFNA was detected in the surface sediment. The concentration of PFOS varied from different depth in the sediment core. The PFOS concentration in the surface sediment varied between 156 pg/g dw to 896 pg/g dw. The concentration decreased with decreasing depth to 6 cm of depth, and after that the concentration varied between 112 pg/g dw to 863 pg/g dw. The

concentration had a peak between 17-20 cm of depth with a PFOS concentration between 1010 pg/g dw to 1630 pg/g dw (Falandysz et al., 2012) 0 50 100 150 200 250 300 350 0 - 2 2 - 4 4 - 6 6 - 8 8 - 1 0 1 0 - 1 4 1 4 - 1 8 1 8 - 2 2 2 2 - 2 6 2 6 - 3 0 CO N CE N TR AT IO N (P G/ G W W ) DEPTH CM

19

In the study from Stockholm Arlanda airport PFOS was the predominant PFAS found in sediment with 86 % of the total PFAS concentration. The highest PFOS concentration (2340 ng/L) was found close to where the fire drills had taken place (Ahrens et al. 2015)

The reason that only PFOS could be detected in the sediment cores indicates a dilution effect, and that other PFAS could instead be in the water phase or the biota (Ahrens et al., 2009). It could also be due to that PFOS has accumulate in the soil and leach out over time. The shorter chain PFASs does not accumulate as well as the long chain PFASs (Buck et al., 2011), which can be a reason for why they are not detected in the core sediment, but in outlet 4. The low sample amount could also be a reason. With a higher sample amount the PFOS concentration that where <LOD could possibly have been detected. There is also a possibility that other PFASs could be detected. Comparing with distribution of other PFASs in Outlet 4 and Vissbäcken, the second most abundant compound had a concentration 3 and 7 % of that of PFOS in those samples. With similar distribution in the lake core sediments these PFASs should be under the detection limit as they are in the samples analyzed.

The increase of PFOS with different depths could be due to decreased pH. Calcium cations could also be a factor of sorption of PFOS in sediment (Ahrens, Lutz et al., 2009; Higgins et al., 2006). It has been suggested that’s electrostatic forces is the reason for this (Higgins et al., 2006)

It can be speculated that the difference in sediment type can have an impact in the result, for example the amount of organic material in the sediment. The sediment from both Vissbäcken and outlet 4 was mostly organic material, while the core sediment was mostly sand in the surface sediment and clay in the deeper sediment. The proportion of organic material in the sample was not determined.

To get a better understanding of the distribution of PFASs in Lake Vänern where Skaraborgs Air command F7 is the point source, samples at a transect from outlet 4 and Vissbäcken might give more information on PFASs distribution in the area.

This study only focus on sediment, but with more water samples a better understanding of the distribution and mobility of the pollutants would have been achieved. The sediment cores will give an historical overview of the usage and expositor of PFAS in the area.

In addition, more sample replicates should have been made for a more reliable result. With a limited sample amount and time, this was not possible.

4.3 PFAS association with metal ions

In another student project cadmium, Cd, antimony, Sb, lead, Pb, and vanadium, V, was detected. Due to low concentrations of Sb, V and PFOS, the samples were recalculated in different units for data to be observed. All samples were in wet weight.

In S-SED1:1 (figure 11) there is an association between increasing and decreasing PFOS, Cd and Sb. The increase of metal ions also increase the PFOS concentration. This could

20

(Ahrens et al., 2009; Higgins et al., 2006). There is a slight association between PFOS and the metal ions in S-SED3:1 (Figure 13), however this contradicts the PFOS peak at 6-8 cm of depth were the metal ion concentration decrease. In S-SED2:1 (figure 12) there is no

significant association between the concentration of PFOS and the metal ions. Between depths 6-22 cm it shows that with an increase in vanadium, the PFOS concentration will decrease. However, cadmium, antimony and PFOS will decrease in concentration at deeper depths (from 10-14 cm depth).

The most significant positive association is between antimony and PFOS where increased antimony increases the PFOS concentration.

To get a more reliable result and get a preferable association between PFASs and metal ions, more PFASs and metal ions should have been detected. Calcium ions and mercury could have been a preferable metal ion to detect due to previous studies. Due to different concentration of all species, an association between PFOS and the metal ions are hard to tell. More work in this field is needed.

For individual associations between PFOS and the metal ions, see figure 21-32, Appendix B.

Figure 11. PFOS and metal ions in sediment core S-SED1:1. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units

0,0 20,0 40,0 60,0 80,0 100,0 120,0 140,0 160,0 0 - 2 2 - 4 4 - 6 6 - 8 8 - 1 0 1 0 - 1 4 1 4 - 1 8 1 8 - 2 2 2 2 - 2 6 2 6 - 3 0 DEPTH, CM Pb (µg/g) Sb (ng/g) Cd (ng/g) V (µg/g) PFOS (pg/g)

21 Figure 12. PFOS and metal ions in sediment core S-SED2:1. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units

Figure 13. PFOS and metal ions in sediment core S-SED3:1. Here, the metals correlates, but not PFOS. Due to variations in concentrations between the different species, the analytes were calculated and reported in different units

0,0 50,0 100,0 150,0 200,0 250,0 300,0 350,0 400,0 450,0 0 - 2 2 - 4 4 - 6 6 - 8 8 - 1 0 1 0 - 1 4 1 4 - 1 8 1 8 - 2 2 2 2 - 2 6 2 6 - 3 0 DEPTH, CM Pb (µg/g) Sb (ng/g) Cd (ng/g) V (µg/g) PFOS (pg/g) 0,0 50,0 100,0 150,0 200,0 250,0 300,0 350,0 0 - 2 2 - 4 4 - 6 6 - 8 8 - 1 0 1 0 - 1 4 1 4 - 1 8 1 8 - 2 2 2 2 - 2 6 2 6 - 3 0 DEPTH, CM Pb (µg/g) Sb (ng/g) Cd (ng/g) V (µg/g) PFOS (pg/g)

22

5.0 Conclusion

The PFAS that significantly dominated in all samples was PFOS. The high PFOS

concentration might be due to extensive use with subsequent release of firefighting foam in the area and degradation of other PFAS substances into PFOS. The largest variation of PFASs was found in outlet 4 with 21 different detected substances, followed by Vissbäcken with 16 different detected substances. The high PFASs concentration might be due to emission of firefighting foam after fire drill exercises at Skaraborgs airport. This could also explain the dominance of PFOS in all sediment samples. Another explanation to the high amount of PFOS could also be by degradation of other PFASs into PFOS.

The only detected PFAS in the sediment cores was PFOS. Some PFOS concentrations were however close to or below the LOD, and was not detected. Despite the emissions from the airport, via outlet 4 and Vissbäcken, not very much PFAS substances accumulate in the accumulation bottom. This may be due to the water solubility of the substances and the large volume of 153 km3.The difference in depth concentrations might be explained by increased pH and calcium ions.

The presence of the second highest PFASs was <3-7% in Vissbäcken and outlet 4. If that is the source and only about 100 pg/g of PFOS was found, 3-7% would have been too low to detect other substances.

There is an association between antimony and PFOS in the sediment. There is also a slight positive association with cadmium.

With a higher sample amount, the PFOS concentrations that were close or under the limit of detection could have been detected. Other PFASs could also have been detected with higher sample amount.

23

6.0 Acknowledgment

I would like to take the opportunity to thank my supervisor Ingrid Ericson Jogsten for all the guide, advice and help during this project work. Whit out her support, I would not been able to finish my work.

I would also like to thank all employees at MTM at Örebro University for all the hospitality and help when needed.

At last, I would like to thank Johan Temnerud for the project opportunity and showing interest, and NIRAS for supplying with samples.

24

References

Anderson, R.H., Long, G.C., Porter, R.C., Anderson J.K. 2016. ‘’ Occurrence of selected perfluoroalkyl substances at U.S. Air Force aqueous film-forming foam release sites other than fire-training areas: Field-validation of critical fate and transport properties’’

Chemosphere, vol. 150 pp 678–685.

Ahrens, L., Norström, K., Viktor, T., Cousins, P.A., Josefsson, S. 2015. ‘’ Stockholm Arlanda Airport as a source of per- and polyfluoroalkyl substances to water, sediment and fish’’ Chemosphere, vol 129 pp 33-38.

Ahren, L., Yamashita, N., Yeung, W.Y.L., Taniyasu, S., Horii, Y., Lam, K.S.P., Ebinghaus, R. 2009. ‘’ Partitioning Behavior of Per- and Polyfluoroalkyl Compounds between Pore Water and Sediment in Two Sediment Cores from Tokyo Bay, Japan’’ Environmental science and technology vol 43, pp 6969–6975.

Banzhaf, S., Filipovic, M., Lewis, J., Sparrenbom, C., Barthel, R. 2017. ‘’A review of contamination of surface-, ground-, and drinking water in Sweden by perfluoroalkyl and polyfluoroalkyl substances (PFASs)’’ Ambio, vol. 46, pp. 335-346.

Bossi, R., Dam, M., Rigét, F.F. 2015. ‘’ Perfluorinated alkyl substances (PFAS) in terrestrial envoronments in Greenland and Faroe Islands’’ Chemosphere, vol. 129, pp 164-169.

Buck, R.C., Cousins, I.T. 2011. ‘’Polyfluorinated Compounds: Past, Present, and Future’’ Environmental science and technology, vol. 45, pp 9821-9821.

DIRECTIVE 2006/122/ECOF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 12 December 2006 amending for the 30th time Council Directive 76/769/EEC on the approximation of the laws, regulations and administrative provisions of the Member States relating to restrictions on the marketing and use of certain dangerous substances and preparations (perfluorooctane sulfonates), L372/32.

Eriksson, U., Kärrman, A., Rotander, A., Mikkelsen, B., Dam M. 2013. ‘’Perfluoroalkyl substances (PFASs) in food and water from Faroe Islands’’, Environmental Science and Pollution Research, vol. 20, pp. 7940-7948.

Falandysz, J., Rostkowski, P., Jarzy´nska, G., Falandysz J.J., Taniyasu, S., Yamashita, N. 2011. ‘’ Perfluorinated carboxylic and sulphonic acids in surface water media from the regions of Tibetan Plateau: Indirect evidence on photochemical degradation’’ Journal of Environmental Science and Health, vol. 47, pp 428-434.

Försvarsmakten (2015e). MTU avseende PFAS inom Skaraborgs flygflottilj, F 7 Såtenäs. 2015-05-13. Rapport – NIRAS Projekt nr. 8114029,8.

Gao, Y., Fu, J., Meng, M., Chen., B., Jiang, G. 2015. ‘’Spatial distribution and fate of perfluoroalkyl substances (PFASs) in sediment from the Pearl River Estuary, South China’’ Marine Pollution Bulletin, vol 96, pp 226-234.

Higgins, P.C., L, G.R. 2006. ‘’Sorbtion of perfluorinated surfactants on sediment’’ Environmental science and technology, Vol 40, pp 7252-7256.

25

Hu, X.C., Andrews, D.Q., Lindstrom,A.B., Bruton, T.A., Schaider, L.A., Grandjean, P., Lohmann, R., Carignan.C, C., Blum, A., Balan, S.A., Higgins, C.P., Sunderland E.M. 2016. ‘’Detection of Poly- and Perfluoroalkyl Substances (PFASs) in U.S. Drinking Water Linked to Industrial Sites, Military Fire Training Areas, and Wastewater Treatment Plants’’ American Chemical Society, vol. 3 pp, 344−350.

Håkansson, Larsson. 1976. ‘’The influence of wind, fetch, and water depth on the distribution of sediments in Lake Vanern, Sweden’’ DOI: 10.1139/e77-040.

Kemikalieinspektionen, 2015 ‘’Occurrence and use of highly fluorinated substances and alternatives’’ Report from a government assignment, report 7/15.

Kissa, E. 2001, Fluorinated surfactants and repellents, 2., rev. and expand edn, Dekker, New York, pp. 29-41.

Kärrman, A., van Bavel, B., Hardell, L., Lindström, G. 2004. ‘’Perfluoroalkylated compounds in whole blood and plasma from the Swedish population’’ Man-Technology-Environment Research Centre dnr 721-4007-02.

lidkoping.se/2015/06/24/pfas-inom-skaraborgs-flygflottilj-f7-satenas/ (2017-05-15, 19.52). Munoz, G., Ciraudel, J-L., Botta, F., Lestremau, F., Dévier, M-H., Budzinski, H., Labadie, B. 2015. ‘’Spatial distribution and partitioning behavior of selected poly- and perfluoroalkyl substances in freshwater ecosystems: A French nationwide survey’’ Science of The Total Environment, vol. 517, pp 48-56.

Naturvårdsverket., Mars 2016. ” Högfluorerade ämnen (PFAS) och bekämpningsmedel en sammantagen bild av förekomsten i miljön”, Redovisning av ett regeringsuppdrag, Rapport 6709, ISBN 978-91-620-6709-0.

Pan, C.G., Ying, G.G., Liu, Y.S., Zhang, Q.Q., Chen, Z.F., Peng, F.J., Huang, G.Y. 2014. ‘’Contamination profiles of perfluoroalkyl substances in five typical rivers of the Pearl River Delta region, South China’’ Chemosphere, vol 114, pp 16-25.

Qi, Y., Hu, S., Huo S., Xi, B., Zhang J., Wang, X. 2015 ‘’Spatial distribution and historical behaviors of perfluoroalkyl substances (PFASs) in sediment of Lake Chaohu, a shallow eutrophic lake in Eastern China, Ecological Indicators, vol. 57, pp 1-10.

Stahl, T., Mattern, D., Brunn, H. 2011. ‘’Toxicology of perfluorinated compounds’’ Environmental Science Europe, DOI: 10.1186/2190-4715-23-38.

Stockholm Convention, 2008. UNEP; Stockholm Convention on Persistent Organic Pollutants; http://chm.pops.int/ (2017-05-29).

Wang, Z., DeWitt, J.C., Higgins, C.P., Cousins, I.T. 2017. ‘’A Never-Ending Story of Polyfluoroalkyl Sub stances (PFASs)?’’ Environmental Science and Technology, vol. 51 pp 2508- 2518.

26

Appendix A

27 Figure. 15 S-SED 2

28 Figure. 16 S-SED 3

29 Figure. 17 Division of sediment samples

30 Figure. 19 Vissbäcken

31

Appendix B

Table 1 List of compounds detected

Compound Abbreviation Molecular formula

PFPeA Perfluoropentanoic acid C5HF9O2

PFBuS Perfluorobutane sulfonate C4F9O3S

PFHxA Perfluorohexanoic acid C6HF11O2

PFHpA Perfluoroheptanoic acid C7HF13O2

PFPeS Perfluoropentane sulfonate C5F11O3S

PFHxS Perfluorohexane sulfonate C6F13O3S

PFHpS Perfluoroheptane sulfonate C6F13O3S

PFOA Perfluorooctanoic acid C8HF15O2

PFOSA Perfluorooctane sulfonamide C8H2F17NO2S

PFOS Perfluoroocatane sulfonate C8F17O3S

PFDA Perfluorodecanoic acid C10HF19O2

PFUnDA Perfluoroundecanoic acid C11HF21O2

PFNS Perfluorononane sulfonate C9F19O3S

PFDS Perfluorodecane sulfonate C10F21O3S

PFDoDA Perfluorododecanoic acid C12HF23O2

PFTrDA Perfluorotridecanoic acid C13HF25O2

PFDoDS Perfluorododecane sulfonate C12F23O3S

PFTDA Perfluorotetradecanoic acid C14HF27O2

PFHxDA Perfluorohexadecanoic acid C16HF31O2

6:2FTS 6:2 fluorotelomersulfonate C8H4F13O3S

8:2FTS 8:2 fluorotelomersulfonate C10H4F17O3S

Table 2 Limit of detection, LOD, for the detected compound

Compound Limit of detection, pg

PFPeA 7,8 PFBuS 1,5 PFHxA 30,3 PFHpA 7,1 PFPeS 0,73 PFHxS 19 PFHpS 0,34 PFOA 13 PFOSA 0,64 PFOS 81 PFDA 14 PFUnDA 17 PFNS 0,98 PFDS 1,1 PFDoDA 37 PFTrDA 37 PFDoDS 0,31 PFTDA 19

32

PFHxDA 37

6:2FTS 74

8:2FTS 1

Table 3. Water content in sediment samples.

Sample Wet weight Dry weight Water content %

S-SED 1:1 0-2 12,34 8,32 33% S-SED 1:1 2-4 13,18 8,28 37% S-SED 1:1 4-6 12,3 7,36 40% S-SED 1:1 6-8 12,21 6,98 43% S-SED 1:1 8-10 12,26 6,67 46% S-SED 1:1 10-14 12,06 6,42 47% S-SED 1:1 14-18 12,15 6,64 45% S-SED 1:1 18-22 12,48 6,7 46% S-SED 1:1 22-26 12,89 6,97 46% S-SED 1:1 26-30 13,57 7,5 45% S-SED 2:1 0-2 12,72 6,2 51% S-SED 2:1 2-4 12,47 9,04 28% S-SED 2:1 4-6 12,26 7,72 37% S-SED 2:1 6-8 12,12 6,57 46% S-SED 2:1 8-10 12,39 7,4 40% S-SED 2:1 10-14 13,17 9,16 30% S-SED 2:1 14-18 12,98 9,3 28% S-SED 2:1 18-22 12,57 8,25 34% S-SED 2:1 22-26 12,47 7,28 42% S-SED 2:1 26-30 12,05 7,47 38% S-SED 3:1 0-2 12,65 5,88 54% S-SED 3:1 2-4 12,8 7,24 43% S-SED 3:1 4-6 12,25 6,85 44% S-SED 3:1 6-8 12,4 6,94 44% S-SED 3:1 8-10 13,22 7,39 44% S-SED 3:1 10-14 12,59 7,14 43% S-SED 3:1 14-18 12,1 6,76 44% S-SED 3:1 18-22 12,02 6,71 44% S-SED 3:1 22-26 12,36 6,97 44% S-SED 3:1 26-30 12,54 7 44%

33 Table 4. PFASs concentrations in water samples from Vissbäcken and outlet 4. The data is from another student project

Compound Vissbäcken, ng/l Utlopp 4, ng/l

PFBA 82,1 22,3 PFPeA 181 52,3 PFBuS 142 58,3 PFHxA 388 113 PFHpA 98,3 28,9 PFPeS 195 71,5 PFHxS 645 397 PFHpS 222 48,1 PFOA 216 53,2 PFNA 8,83 3,78 PFOS 4160 1140 PFDA 0,83 0,93 PFUnDA <LOD 0,28 PFNS 2,01 0,17 PFDS 0,29 0,04 6:2FTS 24,5 48,9 8:2FTS 5,64 <LOD

Table 5. Concentrations Outlet 4 pg/g ww

Compound Concentration pg/g ww PFPeA 221 PFBuS 81 PFHpA 169 PFPeS 113 PFHxA 278 PFHxS 1788 PFHpS 556 PFDA 274 PFUnDA 628 PFNS 104 PFOA 312 PFOSA 1731 PFOS 51772 PFDS 468 PFDoDA 227 PFTrDA 97 PFDoDS 79 PFTDA 142 PFHxDA 44 6:2FTSA 1378

34

8:2FTSA 382

Table 6. Concentration pg/g ww Vissbäcken

Compound Concentration pg/g ww PFPeA 71 PFBuS 16 PFHpA 28 PFPeS 21 PFHxA <LOD PFHxS 335 PFHpS 62 PFDA <LOD PFUnDA 60 PFNS 21 PFOA 59 PFOSA 416 PFOS 7286 PFDS 54 PFDoDA 0 PFTrDA 128 PFDoDS 177 PFTDA <LOD PFHxDA <LOD 6:2FTSA 516 8:2FTSA 118

35 Figure 20. Ranging levels of PFOS in sediment cores showing different sampling sites, depth and values <LOD. PFOS

concentration is calculated in dry weight

Figure 21. Individual association between PFOS and Cadmium in S-SED1:1 -50,0 50,0 150,0 250,0 350,0 450,0 550,0 650,0 0 - 2 2 - 4 4 - 6 6 - 8 8 - 1 0 1 0 - 1 4 1 4 - 1 8 1 8 - 2 2 2 2 - 2 6 2 6 - 3 0 CO N CE N TR AT IO N P G/ G DW DEPTH CM

SEDIMENT CORES

S-SED1:1 S-SED2:1 S-SED3:1 LOD

0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0 20 40 60 80 100 120 140 160 Ca dmi um PFOS

PFOS vs Cd S-SED1:1

36 Figure 22. Individual association between PFOS and Vanadium in S-SED1:1

Figure 23. Individual association between PFOS and lead in S-SED1:1 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 90,0 0 20 40 60 80 100 120 140 160 V PFOS

PFOS vs V S-SED1:1

PFOS vs Pb S-SED1:1

37 Figure 24. Individual association between PFOS and Antimony in S-SED1:1

Figure 25. Individual association between PFOS and Cadmium in S-SED2:1 0,00 0,01 0,01 0,02 0,02 0,03 0,03 0,04 0 20 40 60 80 100 120 140 160 Sb PFOS

PFOS vs Sb S-SED1:1

PFOS vs Cd S-SED2:1

38 Figure 26. Individual association between PFOS and Vanadium in S-SED2:1

Figure 27. Individual association between PFOS and lead in S-SED2:1 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 0 50 100 150 200 250 V PFOS

PFOS vs V S-SED2:1

PFOS vs Pb S-SED2:1

39 Figure 28. Individual association between PFOS and Antimony in S-SED2:1

Figure 29. Individual association between PFOS and Cadmium in S-SED3:1 0,000 0,005 0,010 0,015 0,020 0,025 0,030 0,035 0,040 0,045 0 50 100 150 200 250 Sb PFOS

PFOS vs Sb S-SED2:1

PFOS vs Cd S-SED3:1

40 Figure 30. Individual association between PFOS and Vanadium in S-SED3:1

Figure 31. Individual association between PFOS and lead in S-SED3:1 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 0 50 100 150 200 250 300 350 V PFOS

PFOS vs V S-SED3:1

PFOS vs Pb S-SED3:1

41 Figure 32. Individual association between PFOS and Antimony in S-SED3:1

Table 7. Recovery (%) of internal standards in samples Vissbäcken and outlet 4

Compound Vissbäcken Outlet 4

IS 8:2FTS 48,4 34,4 IS 6:2FTS 48,5 43,4 IS PFBA 89,1 107,1 IS PFHxA 94,3 115,4 IS PFHxS 108,6 108 IS PFOS 98,7 85,4 IS PFOA 88 93,6 IS PFNA NQ NQ IS PFOSA 42,2 23,8 IS PFDA 81,2 76,3 IS PFUnDA 77 68 IS FTDA 47,3 28,8 IS PFDoDA 73,3 65,4 IS PFHxDA 101,5 58,5 NQ: not quantified

Table 9. Recovery (%) of internal standards in samples S-SED1:1

Compound 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 IS 8:2FTS 51,8 51,3 43 39,6 53,5 56,3 43,9 55,9 13,3 24,5 IS 6:2FTS 55,3 60,2 52,5 53,8 64,8 67,8 54,2 68 39 48,7 IS PFBA 62,6 60,9 63,4 57,8 66,3 71,9 62,5 67,7 74,8 81 IS PFHxA 68,2 83,8 71,3 62,6 65,2 87 67,1 83,4 86,6 100,1 IS PFHxS 56,4 64,6 59,2 60,5 73,7 74,7 66,4 70,5 48,8 59,5 0,000 0,005 0,010 0,015 0,020 0,025 0,030 0 50 100 150 200 250 300 350 Sb PFOS

PFOS vs Sb S-SED3:1

42 IS PFOS 56 57,6 59,9 51,2 74,1 72,4 50,7 69,6 21,4 38,3 IS PFOA 43,9 53,1 51,3 52,5 62,5 68,8 57 62,5 39,2 55,4 45,1 52,5 ISPFOSA 35,6 25,2 23,4 25,3 43,1 41,5 27,5 29,4 2,1 6,6 IS PFDA 41,4 43,9 51,3 36,7 57,5 63,5 44 56,3 10 28,4 IS PFUnDA 44,7 43,6 48,8 25,8 53,2 61 33,9 54,7 4,1 15,5 IS PFTDA 98,4 89,4 19,1 2,7 15,8 66 1,3 29,3 0 0 IS PFDoDA 51,1 53,6 33,5 8,7 30,7 42,7 11,1 37,1 0,8 3,4 IS PFHxDA 121,3 102,7 2,2 1,2 0,6 70,8 0,2 3,9 0,1 0 NQ: not quantified

Table 10. Recovery (%) of internal standards in samples S-SED2:1

Compound 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 IS 8:2FTS 80,3 96,1 77 60 26,3 21,9 29,7 26,7 37,9 25,2 IS 6:2FTS 83,7 92,3 78,7 66,3 53,2 46,5 60,6 52,9 63,6 56,7 IS PFBA 78,1 6,4 67,9 69,7 69,2 61,4 78 74,4 80,2 74,2 IS PFHxA 106,5 105,2 94,8 61,6 127,4 121,5 109,5 98,7 154,3 104,3 IS PFHxS 87,7 104,8 86,7 76,7 62,9 51,6 69,8 61,9 74,3 68,1 IS PFOS 88,4 104 88,4 73,9 41 28,7 39,5 38,9 49,9 40,9 IS PFOA 49,7 70,9 50,4 54,6 57,9 48 60,3 56,1 68,8 57,9 IS PFNA 47,9 34,8 47,2 47,2 57,9 45,8 ISPFOSA 53,1 66,7 38,5 21,8 4,1 3,1 5,8 8,4 7,9 6,5 IS PFDA 72,6 69,5 60 62,1 31,6 22,7 30,2 31,8 43,2 29,3 IS PFUnDA 70,7 73,1 79,5 56 21,1 11,2 17,9 17,6 30,2 14,9 IS PFTDA 157,4 99,4 64,3 32,4 0 0,1 0 0 0 0,2 IS PFDoDA 81,4 53,8 40,7 27,4 3,3 3 2,9 0,1 0,1 3,2 IS PFHxDA 607,9 1135,9 183,9 160,2 0 0,2 0 0 0,1 0 NQ: not quantified

Table 11. Recovery (%) of internal standards in samples S-SED3:1

Compound 0-2 2-4 4-6 6-8 8-10 10-14 14-18 18-22 22-26 26-30 IS 8:2FTSA 23 68,3 70,2 28,2 62,4 66,1 70,9 18,4 59,3 38,5 IS 6:2FTSA 50,1 76,1 78,8 44 72,6 74,3 76,6 43,5 69,3 64,6 IS PFBA 67,3 80,6 89,4 52,5 75,2 52,9 78,4 65,7 3,1 79,8 IS PFHxA 106,5 105,3 132,7 74,2 100,1 92,2 88,3 105,3 101,1 105,2 IS PFHxS 58,4 84,8 89,1 52,6 77 80,2 81,8 55,8 79 76,7 IS PFOS 32,1 85,4 79,9 35,9 77,6 76,6 85,4 30,1 74 52,5 IS PFOA 49,7 48,9 45,1 46,7 101,4 56,7 56,4 48,9 87,7 70,9 IS PFNA 37,5 37,8 34 60 ISPFOSA 7,6 14,8 15,1 11,5 21,1 12,3 14 3,5 4,8 13,7 IS PFDA 22,8 66,4 35,5 29,7 17,3 43,5 46,3 18,9 14,9 47,3

43 IS PFUnDA 10,7 61,6 55,1 21,2 44 51,4 49,6 9,2 32,5 30,5 IS PFTDA 0,2 16,6 12,7 0,3 11,7 21,4 10,8 0 6,2 0,1 IS PFDoDA 3,4 11,2 19,7 4,9 18,2 11,1 22,9 1,8 4,1 4,9 IS PFHxDA 0 19,7 11,3 0 10 16,2 9,4 0 5,4 0 NQ: not quantified

Table 12. Results of a reference sample (n=3)

Compound Mean value (pg) Relative standard deviation PFBA 101 43% PFPeA NQ PFBuS NQ PFHxA NQ PFHpA NQ PFPeS NQ PFHxS NQ PFHpS 0,88 77% PFOA 37,5 96% PFNA 13,7 109% PFOSA 0,4 25% PFOS 147 66% PFDA NQ PFUnDA NQ PFNS NQ PFDS NQ PFDoDA NQ PFTrDA 71,5 54% PFDoDS NQ PFTDA NQ PFHxDA NQ PFOcDA NQ 6_2_FTS NQ 8_2_FTS NQ NQ, not quantified