Leaching of Perfluoroalkyl and Polyfluoroalkyl Substances from textiles

(1)

Bachelor Thesis Scheikunde

Leaching of Perfluoroalkyl and Polyfluoroalkyl Substances from textiles

The development and testing of a new leaching protocol

door

Robbert Jan Vos

26 april 2016

Studentnummer

10024409

Onderzoeksinstituut

Verantwoordelijk docent

Instituut Voor Milieuvraagstukken

Pim Leonards & Henk Lingeman

Onderzoeksgroep

Begeleider

(2)

Abstract

Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a large group of compounds with many applications found in mining, paper, electronics coatings and more. PFASs have a very low surface energy which allows them to create a durable water repellency (DWR) effect. The continuous use of these PFASs causes pollution of fluorinated compounds into nature, where they are bioaccumulative and persistent. However, since no substitutes which work as effectively as PFASs have been developed, the phasing out of the usage of PFASs has been going at a slow rate. SUPFES (substitution in practice of prioritized fluorinated chemicals to eliminate diffuse sources) is a research project of universities, industries and knowledge centers whom work together to investigate these fluorinated compounds with the goal to replace these compounds with similar performing compounds without the harmful side effects. As a part of the SUPFES project, a life cycle assessment (LCA) of PFASs containing products needs to be made. A part of this LCA is the analysis of PFASs that leach out of PFASs containing products. While earlier leaching experiments have been performed on different products, no earlier experiments on the leaching of PFASs from textiles have been performed. Therefore, an experimental protocol which quantifies the amount of PFASs which leach from textiles needed to be used. After a successful pilot, the newly developed protocol was used to determine the influence of several variables such as different pH levels, temperatures, leachates and leaching durations on the total amount of PFASs which leach from textiles. These results were compared with the amount of PFASs obtained by a methanol extraction. No relation between pH and leaching of PFASs could be detected. An increase in temperature seemed to increase the leaching rate, but due to a limited amount of samples these results are not conclusive. A clear relation between different leaching time and chain length of these PFASs was observed. Knowing which variables influence leaching rates of PFASs from textiles is a first step in further understanding the leaching process of PFASs from textiles, which in turn may aid in the creation of DWR coatings lacking PFASs compounds.

Populair wetenschappelijke samenvatting

Geperfluoreerde en polyfluoreerde verbindingen (PFASs) zijn binnen de chemie hele bijzondere verbindingen. Deze moleculen hebben unieke eigenschappen die ze erg nuttig maken voor verschillende toepassingen in mijnbouw, electronica, blusschuimen, coatings en vele andere toepassingen. Vooral als coating voor waterafstotende lagen in kleding worden PFASs houdende verbindingen vaak gebruikt. Het wordt vermoed dat deze waterafstotende laag afbreekt, waarbij PFASs moleculen vrijkomen. Helaas zijn deze PFASs schadelijk voor het milieu wanneer deze in de natuur terecht komen. Om deze reden is het wenselijk dat het gebruik van deze gefluoreerde verbindingen afneemt. Als deel van het SUPFES project (substitution in practice of prioritized fluorinated chemicals to eliminate diffuse sources), wat zich bezig houdt met onderzoek naar PFASs en hun alternatieven, is een onderzoeksprotocol opgesteld om de hoeveelheid PFASs wat uit textielen lekt te meten. Met dit protocol is er getest wat de effecten van verschillende pH waarden, temperaturen, vloeistoffen en contacttijd op de hoeveelheid PFASs wat uit textielen lekt zijn. Uiteindelijk bleek dat de pH van het water dat gebruikt wordt weinig invloed heeft. De temperatuur blijkt wel een invloed te hebben. De contacttijd heeft uiteindelijk de meeste invloed op het uitlekken van deze gefluoreerde verbindingen. Deze experimentele procedure en resultaten zijn handig voor vervolgonderzoek, zodat er gemakkelijk vergelijkbare resultaten verkregen kunnen worden, en er nu bekend is welke variabelen duidelijk wel een invloed hebben op de totale hoeveelheid PFASs die uit textielen lekken.

(3)

Table of contents

Introduction ... 4

PFASs and textiles ... 4

PFASs of interest ... 5

PFASs in nature ... 7

SUPFES ... 8

Performed research ... 8

Method and Materials ... 9

Precautions ... 9

Chemicals ... 9

Textiles ... 9

Method ... 10

Leaching experiment procedure: ... 10

Methanol extraction procedure: ... 10

Analysis ... 11

Leaching experiments ... 11

Leaching procedure varying pH ... 11

Leaching procedure varying temperature... 12

Leaching procedure varying leachates ... 12

Leaching procedure varying leaching duration ... 12

Results ... 13 Discussion ... 16 Pilot ... 16 Follow-up experiments ... 16 PH levels ... 16 Temperature ... 16 Leachates ... 16 Methanol extraction ... 17

Leaching over Time ... 17

Conclusion ... 18

Recommendations ... 19

References ... 20

(4)

Introduction

Moving outdoors during rainfall is not something many people enjoy, yet mostly a minor inconvenience. By using an umbrella it is quite easy to prevent getting completely soaked. However, using an umbrella is not always a viable solution. For people working outdoors like lumberjacks or firefighters, sacrificing the use of one hand for keeping yourself dry does not help productivity. For these more specific applications, durable water repellency (DWR) is a viable property for clothing to have. DWR effects are obtained by adding certain coatings to textiles. These additives range from nanoparticles, hydrocarbons, and silicones to side-chain fluorinated polymers (1). Perfluorinated and polyfluorinated compounds (PFASs) are commonly used to form a DWR application on outdoor clothing. Besides outdoor clothing, PFASs are commonly used in applications such as: paper, electronics, mining, pesticides, aqueous film-forming foams and coatings (2).

Despite the structural difference between these DWR additives, the DWR effect is created equally for all of them: The DWR effect is the result of the differences between the surface energy of the contact area (the textiles in this case), and the surface tension of liquids which come in contact with the textile (2). Table 1 shows the surface energies and surface tensions for some common functional groups and liquids.

Table 1: Surface energies and tensions for commonly used materials and molecules (2)

Surface Liquids Surface energy: γc (mN/m) Surface tension: γL (mN/m) –CF3 6 –CF2H 15 –CF2– 18 –CH3 22 –CH2– 31 –CH2CHCl– 39 Polyester 42 Polyamide 46 Cotton 44 Water 72 n-Octane 22 Olive oil 32

It is clear from table 1 that the surface energy of the fluorinated compounds are lower than the hydrocarbon counterparts. The lower surface energy results into the increased effectiveness of the DWR application. These fluorinated compounds surface energies are even low enough to provide an oil repellency effect as well.

PFASs and textiles

While the methods of applying a DWR property to textiles are industrial secrets, these methods are similar enough to be explained broadly (2, 3). After weaving, knitting or otherwise forming the basic textiles, these

(5)

textiles undergo finishing treatments. These treatments consist of steps like dyeing, printing, or adding different additives. Adding a DWR layer made up of PFASs is one of such additives. These DWR compounds are added by a so called pad-dry-cure process in which the fabricated textile is passed through an aquatic solution containing several PFASs and possibly cross-linkers. This step is followed up by squeezing the remaining liquids out of the fabricated textiles with pressure pads. This textile is then dried and cured in an oven with temperatures up to 180 °C. It is in this oven the added PFASs starts to polymerize and crosslink with each other. This crosslinking is important for the durability of the DWR layer (4). Despite the differences in structure between different types of DWR layers, the structures of the created PFASs polymers which provide the DWR effect are roughly the same. The polymerization and crosslinking between the monomers can be performed with monomers containing either carbon or non-carbon atoms like silicon. The monomers bind with each other, either by an ether bond, or a direct bond between the carbon or silicon backbone of the monomers. The monomers contain perfluorinated or polyfluorinated sidechains which form a DWR effect. A commonly used group of sidechains are the so called fluorotelomers, which will be discussed later (2).

While the polymers are stable, degradation of the effectiveness of the DWR layer has been noticed (5, 6). This means that at least some form of PFASs are released from the textiles which results into this diminishing of effectiveness. The exact group of molecules which leach out of the textiles depends on the molecules used to produce the polymers. For example, the commonly used fluorotelomer alcohols (FTOHs) degrade to perfluoroalkyl carboxylic acids (PFCAs), while fluorotelomer sulphonic acids (FTSAs) are known to degrade to perfluoroalkane sulphonic acids (PFSAs) (2). The degradation of these polymers are caused by several factors. The most notable factors are hydrolysis, photo-oxidation and UV degradation (2). These factors might cause the sidechains of the polymers to break off and to be released into the nature. Another source of PFASs which are released in nature are pollutions which originate from the creation of the DWR layer. These are the monomers used in forming the DWR application, and PFASs pollutants found in these monomers.

PFASs of interest

Considering fluorinated compounds have been in large scale production for over 50 years, many different names and abbreviations have been used to name all these compounds. Buck et all. (7) did an excellent job trying to standardize all the different nomenclature. The research discussed in this report focusses on fluorotelomers and perfluorinated acids. Using Bucks nomenclature, figure 1 showing relevant PFASs groups and subgroups was created. Table 2, 3 and 4 show the structure, acronyms, cas numbers and names of several relevant compounds to this research.

(6)

(7)

Table 2: Names, structures and cas numbers of common perfluoroalkyl carboxylic acids (PFCA)

Name Acronym Structure Cas number

Perfluorobutanoic acid PFBA CF3-(CF2)2 -CO2H 375-22-4

Perfluoropentanoic acid PFPeA CF3-(CF2)3 -CO2H 2706-90-3

Perfluorohexanoic acid PFHxA CF3-(CF2)4 -CO2H 307-24-4

Perfluoroheptanoic acid PFHpA CF3-(CF2)5 -CO2H 375-85-9

Perfluorooctanoic acid PFOA CF3-(CF2)6 -CO2H 335-67-1

Perfluorononanoic acid PFNA CF3-(CF2)7 -CO2H 375-95-1

Perfluorodecanoic acid PFDA CF3-(CF2)8 -CO2H 335-76-2

Perfluorundecanoic acid PFUnDA CF3-(CF2)9 -CO2H 2058-94-8

Perfluorododecanoic acid PFDoDA CF3-(CF2)10 -CO2H 307-55-1

Perfluorotridecanoic acid PFTrDA CF3-(CF2)11 -CO2H 72629-94-8

Perfluorotetradecanoic PFTeDA CF3-(CF2)12 -CO2H 376-06-7

Table 3: Names, structures and cas numbers of common perfluoroalkyl sulfonic acids (PFSA)

Perfluorobutane sulfonic acid PFBS CF3-(CF2)3 -SO3H 375-73-5

Perfluorohexane sulfonic acid PFHxS CF3-(CF2)5- SO3H 355-46-4

Perfluoroheptane sulfonic acid PFHpS CF3-(CF2)6 -SO3H 375-92-8

Perfluorooctane sulfonic acid PFOS CF3-(CF2)7 -SO3H 1763-23-1

Table 4: Names, structures and cas numbers of common fluorotelomer sulfonic acids (FTSA), fluorotelomer carboxylic acids (FTCA) and fluorotelomer alcohols (FTOH)

4:2 Fluorotelomer sulfonic acid 4:2 FTSA CF3-(CF2)3-(CH2)2-SO3H 757124-72-4

6:2 Fluorotelomer carboxylic acid 6:2 FTCA CF3-(CF2)5-CH2-CO2H 53826-12-3

6:2-Fluorotelomer alcohol 6:2 FTOH CF3-(CF2)5-(CH2)2-OH 647-42-7

PFASs in nature

Earlier it was mentioned that PFASs used in DWR applications are released from these textiles. These PFASs end up in nature causing several negative effects. While the exact nature of the degradations of

(8)

PFASs containing DWR layers are not clearly defined yet (4), the effects of PFASs pollution have been researched. The Madrid statement on Poly- and Perfluoralkyl Substances (8) gives a clear summary of several negative effects from PFASs exposure. This summary shows some long-chain PFASs can cause tumors in multiple organ systems, toxicity for newborns possibly resulting in death, different neurobehavioral effects, disruptions of the immune, endocrine and lipid metabolism systems and liver toxicity. On top of these effects, PFASs exposure has been associated with more negative effects such as: Testicular and kidney cancers, liver malfunction, hypothyroidism, high cholesterol, ulcerative colitis, lower birth weight and size, obesity, decreased immune response to vaccine and reduced hormone levels and delayed puberty. The reasons PFASs are useful as DWR (they are stable under extreme conditions, strong bonds, inert) also causes some of the negative effects in nature (2). Because of the strong carbon-fluor bonds, these PFASs barely degrade in nature. On top of the fact PFASs are persistent, they also happen to be bioaccumulative, meaning these PFASs will accumulate in organisms (9). Fortunately, some steps have been take to prevent PFASs pollution. The Stockholm Convention (10) declared PFOS to be a Persistent Organic Pollutant (POP), resulting in a ban of production of this compound. Currently it is being discussed whether PFOA should join PFOS as a POP (2). A maximum level of PFOS in textiles has been determined by the European Union to be 1 μg/m2 _{(11). Due to this ban, PFOS is no longer used in current applications}

of DWR. This resulted in usage of different PFASs not yet banned in the EU, which share some of the same negative attributes when compared to PFOS (2).

SUPFES

Because of the increased awareness of the harmful effects of PFASs in nature and the ban of using PFOS, even more research into these compounds needs to be performed. Whether it are industries searching for better alternatives, or universities investigating the effects of exposure to PFASs, the need for more knowledge arises. Because of this reason the SUPFES (Substitution in Practice of Prioritized Fluorinated Chemicals to Eliminate Diffuse Sources) project has been founded. The focus of SUPFES is studying emissions, life cycles and human and aquatic toxicity connected with DWR in textiles (12). As a part of a life cycle assessment (LCA) focused specifically on PFASs, a quantification of the amount of PFASs which leach from textiles during daily usage is required. While extractions to measure the total amount of PFASs in textiles have been performed (4, 13), no leaching experiments for PFASs out of textiles have yet been performed. Because of this lack of research on the leaching of PFASs out of textiles, a new protocol needs to be developed.

Performed research

In this research, a partially new protocol was described which was used to quantify the amounts of PFASs which leached out of a piece of textile. The leaching experiment was created ourselves, while the analysis of the watery leachate for the concentration of PFASs was based on an existing protocol. This protocol made use of solid phase extractions (SPE) to extract the PFASs from the leachate.

A different, second extraction was used which consisted of a liquid-solid extraction (LSE) with methanol to extract PFASs directly out of textiles (13).

The effects of different variables such as pH levels, temperatures, leachates and leaching durations were researched.

(9)

Method and materials

Precautions

For all experiments performed in this research some specific guidelines need to be followed: No glass or Teflon materials could be used since it is known that PFASs is adhesive to glass, while Teflon is known to leach PFASs passively. Since household dust contains PFASs (4), extra care had to be taken to prevent household dust to contaminate the samples by working in a fume hood when handling the textiles and rinsing of the textile to wash away dust. Internal standard was added to the samples to correct for losses during extraction steps and to correct for matrix effects with LC-MS analysis.

Chemicals

All isotope-labeled PFAAs were purchased from Greyhound Chromatography (Merseyside, UK) in solutions of 50 mg/mL in methanol. The purity was >98% The isotope purity of 18_O

2-PFHxS was >94%,

and the isotope purity of all other isotope-labeled PFAA was >99%. All used ultrapure water was supplied from a Mili-Q system from Milipore (Watford, UK). Filtering of the mobile phase was performed with glass fiber filters (GF/F, poresize 0.42 μm) purchased from Whatman (Maidstone, UK). Nitric acid (65%, for trace analysis) and Potassium hydrogen phthalate (≥ 99.5, for analysis) were produced by Merck. Sulphuric acid (95%-97%, ACS reagent), acetic acid (> 99.8%, ACS reagent), ammonium acetate (≥ 98.0%, Reagent) and ammonium hydroxide solution (reag. ISO) were produced by Sigma Aldrich. Hydrochloric acid (36.5%-38%, trace metal analysis) and methanol (Ultra HPLC Grade) were produced by J.T Baker. Sodium Hydroxide (1M, 1.000N), potassium chloride (≥ 99.5%, FLUKA guarantee) and sodium bicarbonate (≥ 99.5%) were produced by FLUKA. Glycine (≥ 99.0%, Bio Ultra) and Potassium phosphate monobasic (≥ 99.0%) were produced by Sigma Life. Tris(hydroxymethyl)aminomethane (Ultrapure) was produced by Applichem. More information about the used chemicals can be found in table A1 in the appendix.

Textiles

All textiles used by SUPFES were supplied from several industries known to produce PFASs treated textiles. Of these textiles, two different pieces of textiles were used: A yellow and a pink piece of textile. The yellow piece of textile was used during a pilot experiment which was performed with the goal to determine whether any quantifiable amount of PFASs would leach from a piece of textile. After a positive result, the follow up experiments were performed on the pink piece of textile due to a shortage of the yellow textile. The pink textile was used in both the methanol extractions and the leaching experiments. No methanol extraction was performed on the yellow piece of textile. Samples of approximately 25 cm2 _{were used for the pilot}

experiment with the yellow textile. All follow up experiment were performed on the pink textile on pieces with an area of approximately 16 cm2_{. For both textiles, the mass and surface of a larger piece of textile}

were used to determine the area/mass ratio. By measuring the mass of the samples, corrections were made for deviations from the targeted sample area of 16 cm2_.

(10)

Method

Leaching experiment procedure:

The conditions and experimental procedure which were used to leach PFASs from textiles were chosen ourselves, while the analysis of the leachate was based on a protocol for PFASs analysis available at the IVM, and on information available in literature (14). A piece of textile, with a known surface area, was rinsed from contaminants by submerging the textile once in ultrapure water. The piece of textile was added to a 50 mL polypropylene tube. The polypropylene tube was filled to 45 mL with a specific leachate and shaked with a shaking machine for a specific time. After leaching, the textile was removed and stored in a clean polypropylene tube for later extraction. Internal standard was added to the leachate. To concentrate and extract the PFASs, SPE using an OASIS weak anion exchange (WAX) cartridge was used. The WAX cartridges were conditioned to ensure optimal wetting of the cartridge by the water based leachate. Conditioning was performed by adding 4 mL 0.1% ammonium hydroxide in methanol, followed up by 4 mL methanol. A third conditioning liquid made up of 4 mL ultrapure water was also run through the cartridge. These steps were performed in a way which ensured the SPE cartridge did not run dry, by closing the SPE cartridge using a small tap in which the SPE cartridge was put. After conditioning the SPE cartridge, the leachate was added to the SPE cartridge which resulted in an elution rate of 1 drop per second. This elution rate was controlled by applying vacuum to the SPE cartridge standard. After all leachate was added, the polypropylene tube was washed with ultrapure water and this water was also run through to the SPE cartridge. After all liquids containing any PFASs were added to the SPE cartridge, the cartridge was washed with a 25 mM acetic acid buffer with pH 4.0. Afterwards, vacuum was applied to dry the SPE cartridge. Neutral PFASs compounds were extracted from the cartridge with 4 mL methanol (fraction I), followed by extraction of ionic PFASs with 4 mL 0.1% NH4OH in methanol (fraction II). Fraction I was stored for

future analysis.

The methanol of fraction II was evaporated to dryness under a gentle stream of nitrogen in a water bath at 40°C. 200 µL of a water/methanol mixture (1:1 v/v) was added to the polypropylene tube. The tube was centrifuged at 3000 RPM for 10 minutes to ensure any solid particles remained in the polypropylene tube while the liquid containing the PFASs was pipetted to a 300 µL vial. This vial was put in a LC-MS for analysis.

Methanol extraction procedure:

For the methanol extraction to determine the total amount of PFASs, a protocol described by Veen et al. was used (16). A textile sample with an area of 16 cm2_{was obtained. The piece of textile was rinsed from}

dust particles by dipping the textile shortly in ultrapure water. The piece of textile was put into a 15 mL polypropylene tube. Internal standard was added to the textile. After waiting for an hour in which the internal standard was adsorbed to the textile, 7 mL methanol was added to ensure that the textile was completely submerged in methanol. The polypropylene tube was vortexed and agitated by a shaking machine on 300 RPM for 30 minutes. The methanol was transferred to another polypropylene tube, and a second batch of methanol was added to the polypropylene tube which contains the textile. After vortexing, the polypropylene tube was agitated again by a shaking machine on 300 RPM for 30 minutes. The second batch of methanol was added to the first batch and the polypropylene tube containing the textile was discarded. The methanol was evaporated under a gentle stream of nitrogen in a water bath at 40 °C. 200 µL of a water/methanol mixture (1:1 v/v) was added to the polypropylene tube. The tube was centrifuged on 3000 RPM for 10 minutes to ensure any solid particles remained in the polypropylene tube while the

(11)

water/methanol mixture containing the PFASs was pipetted into a 300 µL polypropylene vial. This vial was put in a LC-MS for analysis.

Analysis

Analysis of all samples were performed by a Agilent 6410 Triple Quad LC-MS/MS (Agilent Technologies, Amstelveen, The Netherlands) using electrospray negative ionization. A symmetry C18 guard column

(20mm x 3.9 mm, 5 µm; Waters Corporation, Milford, Massachusetts, USA) combined with a

FluoroSep-RP Octyl column (150 mm length x 2.1 mm i.d., 5 µm; ES industries, Bellefonte, Pennsylvania, USA) was

used to perform separation. An extra column was installed (Symmetry C18, 150 mm length x 2.1 mm i.d., 5

µm; waters corporation) between the pump and auto-sampler to retain PFASs leaching out the HPLC and mobile solvents. Two mobile solvents were used; A 5 mM ammonium formate solution, filtered over glass fiber filters (GF/F, pore size 0.42 μm; Whatman, Maidstone,UK) and methanol. The injection volume was 20 µL. In Table 5, the gradients of the PFASs analysis are given.

Table 5: Gradients of PFAS analysis on LC-MS/MS

Time (min.) 5 mM Ammonium formate (%) Methanol (%) Flow rate (µL/min.)

0 65 35 300 1 65 35 300 1 05 95 300 21 05 95 300 21.2 05 95 400 27 05 95 400 27.5 65 35 400 33.5 65 35 400 34 65 35 300

Quantification of all extracts was performed against five calibration samples with concentrations of 30 ng/mL, 5 ng/mL, 1 ng/mL, 0.15 ng/mL and 0.05 ng/mL, which were prepared from a single stock solution. All results were quantified using the program “Quantitative analysis for QQQ” of Agilent Technologies, which is a part of Masshunter Workstation software. Quadratic curves and a curve fit weight of 1/x, with x being the relative analyte concentration, were used. Procedure solvent blanks were also analyzed alongside the samples and subtracted from the final results. The limit of detections (LOD) was determined to be three times the noise signal divided by the sample intake and the recovery. The limits of quantification (LOQ) were calculated to be 10 times the noise signal. Table A2 in the appendix shows all mass transitions used during analysis.

Leaching experiments

Leaching procedure varying pH

To determine the effect of different pH levels on the leaching of textiles, several buffered leachates were prepared. The pH levels which were tested were: pH 2, pH 3.5 pH 4.5 pH 5.0 pH 7.0 pH 8.5 and pH 10 to ensure a broad range of pH levels which a piece of textile might be exposed to. Information on the used

(12)

chemicals for these buffers can be found in table 6. Leaching was performed by the previously mentioned protocol.

Table 6: Chemicals used for buffering the leachates

pH Salt Used salt (g) Added acid/base Filled to

(mL)

2.0 Potassium Chloride 0.1711 0.585 mL 1M HCl 45

3.5 Glycine 0.4575 0.360 ml 1M HCl 45

4.5 potassium hydrogen phthalate 0.4593 0.380 mL 1M NaOH 45 5.0 potassium hydrogen phthalate 0.4632 1.020 mL 1M NaOH 45 7.0 Monopotassium phosphate 0.3920 1.655 mL 1M NaOH 45 8.5 Tris ((hydroxymethyl) aminomethane) 0.2755 0.640 mL 1M HCl 45

10 Sodium bicarbonate 0.0952 0.480 ml 1M NaOH 45

Leaching procedure varying temperature

To determine the effects of different temperatures on the leaching of textiles, two pieces of textiles were prepared with ultrapure water as leachate. Of these two samples, one was stored in a stove with a temperature of 40 °C while the second sample was stored at lab temperature of 20 °C. These samples were not constantly agitated since the shaking machine could not be combined with an increased temperature. Manual shaking was performed approximately 20 times during the leaching time, by taking the polypropylene tubes and shaking these tubes firmly by hand for 5 seconds.

Leaching procedure varying leachates

To determine the effects of different leachates on the leaching of textiles, several different leachates were prepared. Different leachates like (synthetic) rainwater and detergents were prepared. Rainwater was collected on the 14th_{of January 2016 in Huizen during the night. Synthetic rainwater based on EPA protocol}

1312 was prepared using nitric and sulphuric acid (15). A buffer with pH 5.0 was prepared, based on the acidity of rainwater which was determined to be pH 5.0. A leachate of “Color Reus Bonte Was” prepared by following the recipe of a hand wash contained on the packing, which noted 60 mL of detergent per 10 liters of water. A “NIKWAX tech wash” mixture, used to restore a PFASs based DWR layers, was prepared by using the hand wash recipe as found on the Nikwax website. 2.5 mL of Nikwax was diluted in 1L of ultrapure water. These samples were leached for a week.

Leaching procedure varying leaching duration

To determine the effects of time on leaching, leaching of textiles with different durations were performed. By removing the textiles which were being leached, an overview of the effects of time on leaching quantities could be observed. The following leaching durations were used: 1 hour, 4 hours, 1 day, 2 days, 4 days, 8 days, 16 days and 21 days. To ensure constant conditions during these different leaching times, a buffered solution with pH 5.0 was used as a leachate. This will prevent the dissolving of CO2in the leachate to

(13)

Results

Figure 2 and 3 show the detected concentrations of PFASs during the pilot experiment. Figure 2 shows the concentrations of PFCA, while figure 3 shows the concentrations of FTCA. Shown in figure 4, 5, 6 and 7 are the results of all experiments performed with the goal to determine the influence of several variables, which were performed after the successful pilot. Figure 4 shows the concentrations by varying pH levels. Figure 5 shows the concentrations when leaching was performed at different temperatures. Figure 6 shows the concentrations when leaching was performed with different leachates. Figure 7 shows the concentrations of PFCA when extracted with methanol. Figure 8 shows the concentrations when leaching was performed with different durations. Figure 9 shows the concentrations of PFCA when the textiles were extracted with methanol after leaching over time was performed. Table A3 in the appendix shows the concentrations of all compounds detected. Table A4 shows the recoveries of the internal standards which were added to the samples.

Figure 2: PFCA detected in the pilot experiment

Figure 3: FTCA detected in the pilot experiment 0 1 2 3 4 5 6 7 8 9

PFBA PFPeA PFHxA PFHpA PFOA PFNA PFDA

Con ce n tra tio n (µ g/m 2)

PFCA found in pilot

Synthetic Acidic Rain Diluted Hydrochloric Acid Rainwater

0 20 40 60 80 100 120 140

6:2 FTCA 8:2 FTCA 10:2 FTCA

FTCA found in pilot

(14)

Figure 4: Concentrations by varying pH levels

Figure 5: Concentrations at different temperatures

Figure 6: Concentrations with different leachates 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8

PFCA concentrations varying pH levels

pH 2.0 pH 3.5 pH 4.5 pH 5.0 pH 7.0 pH 8.5 pH 10 0 0,1 0,2 0,3 0,4 0,5 0,6

PFCA concentration varying temperatures

Leaching at 293K Leaching at 313K 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8

PFCA concentrations varying leachates

(15)

Figure 7: Found concentrations with methanol extraction

Figure 8: Concentrations with varying leaching duration

Figure 9: Found concentrations with methanol extraction after leaching over time 0 0,2 0,4 0,6 0,8 1

PFBA PFPeA PFHxA PFHpA PFOA PFNA PFDA PFUnDA PFDoDA

Total PFCA concentrations

Methanol Extraction 1 Methanol Extraction 2

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7

PFCA leaching concentrations over time

1 Hour 4 Hour 1 Day 2 Days 4 Days 8 Days 16 Days 21 Days

0 0,1 0,2 0,3 0,4 0,5 0,6

PFBA PFPeA PFHxA PFHpA PFOA PFNA PFDA PFUnDA PFDoDA

Methanol extraction after leaching over time

(16)

Discussion

Pilot

The main objective of the pilot experiment was to determine whether leaching was observable and quantifiable using our protocol which consisted of using 45 mL of leachate and performing leaching for a week. When we look at figure 2 and 3, we clearly see that PFASs leach out of this textile when using these variables. Because this experiment was only a pilot, a method blank was ignored. Therefore, accurate quantification for these experiments was not possible. Despite the lack of accurate data, we still see concentrations significantly higher than the limit of PFOS imposed by the EU (1 μg/m2_{). The amount of}

PFOS which leached out the textile during the pilot was factor 2 to 3 times higher than allowed. When looking at PFHxA, we see leaching up to 7 μg/m2_{. The concentrations for 6:2 FTCA were very high. No}

approximations can be given due to the fact the found signal exceeded the calibration curve. Follow-up experiments

PH levels

When looking at the results found in figure 4, we see that several samples lack a concentration for PFBA. When looking at the recoveries of these samples, we see that for these samples barely any internal standard was recovered as well. Therefore, nothing conclusive can be said about any PFBA concentrations.

Looking at figure 4, it appears that leaching at a pH level of 7.0 is less prevalent than on other pH levels. However, a leaching experiment performed on PFASs treated carpets showed no significant changes in leaching rate between pH levels of 5, 6, 7 and 8 (16). Considering the pKa levels of PFAA are <1, no obvious explanation for this result can be given. There do not appear to be any clear effects on the leaching rate of PFASs from textiles with varying pH levels. Repeating the experiments for more data points may result in clearer to interpret results.

Temperature

Looking at figure 5, we can see a higher leaching rate for PFPeA, PFHxA and PFOA at 313K while a lower leaching rate for PFHpA was found at 313K. Only for PFPeA and PFHxA were the signals actually above LOQ. The signals of PFBA, PFHpA, PFOA and PFNA were between LOD and LOQ. Considering values between LOD and LOQ are to be avoided, it is hard to conclude any trends with only two data points. Earlier research on carpets showed an increase of leaching from carpets with an increase in temperature (16). While this conclusion seems to fit the experimental results, it would be better to repeat the experiment multiple times for more accurate results.

Leachates

Looking at figure 6, we see no clear different in leaching effects between Nikwax and Color Reus detergent. What was remarkable however, was the extremely low recoveries which were observed with the Color Reus detergent sample and blank. Lower than average recoveries were observed for any PFCA with carbon chains longer than or equal to PFHpA. A possible explanation for this might be some active components are included in the detergent, which absorb and remove some PFASs found in the leachate. Another explanation may be a matrix effect of different components found in the detergent.

(17)

The NIKWAX tech wash was included because of the claimed regenerative properties to the DWR layer of PFASs treated coatings. Because of these claims it might be possible different results were to be obtained when textiles were leached either Nikwax or other non-regenerative detergents. When looking at figure 6, we see similar leaching for both Nikwax and Color Reus detergents. No differences between these two detergents were found.

Because the recovery of internal standard of PFBA was very low, nothing can be said about this specific component. The difference between rainwater and leaching in water brought to pH 5.0 with hydrochloric acid is interesting. Less leaching appears to happen in the sample of rainwater compared to a buffer with similar pH. A possibility is the presence of cations which can create a bridge between negatively charged PFASs and other negatively charged surfaces (16).

Methanol extraction

The methanol extraction was proved to be a valid method to extract all the PFASs pollutants which were still present on the textile (13). The reported relative standard deviations were <9% for repeatability, and < 20% for reproducibility. Looking at the results in figure 7, we see some deviations in concentrations between each textile. While the spread between the found concentrations is quite large, it does not exceed the 20% found for reproducibility.

Leaching over time

Because the recovery of internal standard of PFBA was very low, no quantified data can be given on this compound. The recoveries for leaching with a duration of 1 day are lower since some leachate was spilled. See table A4 for the recoveries of all samples. Looking at figure 8 and 9, we see a clear connection between leaching and the time in which the textile was brought in contact with the leachate. Figure 8 shows a clear increase in found concentrations for PFCA when leaching was performed for a longer time. Figure 9 shows a clear decrease in remaining PFCA which could be found in textiles after leaching was performed. When combine figure 8 and 9 we get figure 10. Here we see the total amount of PFCA found split between the part found by leaching and the part found after methanol extraction expressed in percentage compared to the average of the two methanol extractions. This figure shows us that for the most part, the leaching and methanol extraction together add up to a total amount of PFCA which is approximately equal for each compound. The major outliers are the concentrations of PFOA found on the 16 and 21 days of leaching.

Figure 10: Total amount of PFAA found with leaching and methanol extraction, expressed in percentage 0 20 40 60 80 100 120 2h 2d 8d 21d 1h 1d 4d 16d 2h 2d 8d 21d 1h 1d 4d 16d 2h 2d 8d 21d 1h 1d 4d 16d

PFPeA PFHxA PFHpA PFOA PFNA PFDA

To ta l L each in g (% )

Combined leaching and methanol extraction (%)

(18)

Figure 10 also shows a decrease of contribution of the leaching over time to the total amount of PFASs found on the textile. When the length of the PFASs increase, the solubility of those compounds decrease. PFNA and PFDA both barely leach out of the textiles, and most of the concentrations of those compounds were recovered during the methanol extractions. This is not the case for the PFCAs with a chain length of 5 to 8 carbons.

Another interesting aspect of the leaching over time is seen when the leaching over time is expressed in percentage compared to the total amount of leaching possible, based on the average of the methanol extractions. Figure 11 shows the graph which shows the total amount of leaching over time per compound. One could expect to see a correlation between the length of the PFCA and the rate of observed leaching. When looking at figure 11, we see this correlation to be true. The form of the graph also shows a smoothing of the curve indicating a faster leaching rate for compounds up to 8 perfluorocarbons long. For compounds longer than 8 perfluorocarbons, more time is needed before this flattening could possibly be observed.

Conclusion

This research started with the goal to test for several possible variables which may influence leaching rates of PFASs from textiles. This in turn required a new protocol since no leaching experiments had been performed on pieces of textile before. When looking at the results of the pilot and the results of the second piece of textile, it is clear leaching effects can be quantified using our new method.

This new leaching protocol has been used multiple times on different pieces of textile with different variables. While no data has been collected on repeatability and reproducibility, this method appears to be quite robust except for the shortest perfluorinated compounds.

Using this developed protocol, some possibly relevant variables on the leaching rate have been examined. Some of these tested variables showed no significant result, others showed an inconclusive difference, while other variables showed a clear influence on the leaching rate.

While a lower leaching rate was observed for a pH of 7, other leaching experiments testing pH levels showed no deviations. Considering the fact each pH level used a different buffer, not all variables were equal and no definitive conclusion can be given for the influence of pH.

Leaching under different temperatures seemed to show a variable leaching rate, despite the fact not all analyzed components followed this trend.

The testing of different leachates by using two detergents and rainwater was not conclusive when considering the small sample size. Leaching for different durations did show some useful and convincing results.

It was clear that time of leaching had an effect on the amount of PFASs which were leached from the textile. When expressing these concentrations in percentage, the rate of leaching seemed to correlate almost directly with the chain length of the analyzed PFASs. When the textile residues from each of these leaching experiments over time were extracted from all remaining PFASs by methanol extraction, the total amount of PFASs which could be found on each sample seemed to be quite similar to each other.

If researchers decide to use this protocol to determine leaching effects on textiles, it is recommended to keep the time of leaching and temperatures equal at all times. While differences in pH levels did not show

(19)

a clear relation, differences in leaching between samples on different pH levels was observed. It is encouraged to experiment with buffered solutions to keep this variable as equal as possible.

Recommendations

Some experiments, like measurements of differences between temperature and methanol extractions, were performed on a very small number of samples. Having performed these measurements on more samples would have increased the precision of this experiments. Most experiments assumed an equal concentration of PFASs on each sample of textile. If this were the case, either small or no differences should have been observed between the two methanol extractions.

Another recommendation would be a re-analysis of all samples to correct any errors made due to inexperience of the researcher performing these experiments. Especially deviations from the protocol when working with internal standard could cause significant deviations in the found concentrations.

While several experiments have been performed, another experiment was planned. A follow up experiment in which rainfall would be simulated was developed, but not executed due to time restraints. During these leaching experiments, the textiles were fully submerged in a polypropylene tube. When wearing outdoor clothing during rainfall, rainfall does not fully submerge the outdoor clothing. Having a small scale simulator which would allow leachate to fall on the textile with the orientation of the PFASs hydrophobic layer to be correct could possibly show different leaching results compared to full submerging of textiles. Current estimates of leaching rates cannot be made since these experiments were developed for easy reproducibility, not for precise replication of environmental variables.

(20)

References

[1] Holmquist, H. et al. Properties, performance and associated hazards of state-of-the-art durable water repellent (DWR) chemistry for textile finishing. Environment International. 2016, 91, 251-264

[2] Knepper, T. P. et al. The Handbook of Environmental Chemistry 17. Springer, Berlin, 2012

[3] Mukhopadhyay, A. et al. A review on designing the waterproof breathable fabrics part I: fundamental principles and designing aspects of breathable fabrics. J. Ind. Text. 2008, 37, 225–262.

[4] Knepper, T. P. Understanding the exposure pathways of per- and polyfluoralkyl substances (PFASs) via use of PFASs-Containing products risk estimation for man and environment. 2014. Umweltbundesam, Dessau-Roßlau

[5] Arunyadej, S. et al. An investigation into the effect of laundering on the repellency behaviour of a fluorochemical-treated cotton fabric. J. Text. Inst, 1998, 89, 696–702

[6] Leonas, K.K., Effect of laundering on the barrier properties of reusable surgical gown fabrics. Am. J. [7] Buck, R. C. et al. Perfluoroalkyl and Polyfluoroalkyl Substances in the Environment: Terminology, Classification, and Origins. Integr.Environ.Assess.Manag, 2011, 7, 513–541

[8] Blum, A. et al. The Madrid statement on poly- and perfluoroalkyl substances (PFASs). Environ Health Perspect 123:A107–A111, 2015,; http://dx.doi.org/10.1289/ehp.1509934

[9] Lau, C. et al. Perfluoroalkyl acids: A review of monitoring and toxicological findings. Toxicol. Sci, 2007, 99 (2), 366-394

[10] UNEP, United Nations Environment Programme, Stockholm convention on persistent organic pollutants: guidance on alternatives to perfluorooctane sulfonic acid, its salts, perfluorooctane sulfonyl fluoride and their related chemicals, UNEP/POPS/POPRC.9/INF/11/Rev.1, Rome, 18 October 2013, 1-63 〈http://chm.pops.int/TheConvention/POPsReviewCommittee/Meetings/

POPRC9/Documents/tabid/3281/Default.aspx〉.

[11] EU, Directive 2006/122/EC of the European Parliament and of the Council of 12 December 2006, Off. J. Eur. Union L372/32 (2006) 32–34 http://eur-lex.europa.

eu/LexUriServ/LexUriServ.do?uri¼OJ:L:2006:372:0032:0034:en:PDF.

[12] SUPFES homepage. http://www.supfes.eu/ProjectInfo.aspx (accessed on 26-04-2016)

[13] van der Veen, I. et al. Development and validation of a method for the quantification of extractable perfluoroalkyl acids (PFAAs) and perfluorooctane sulfonamide (FOSA) in textiles. Talanta, 2016, 147, 8- [14] Taniyasu, S. et al. Analysis of fluorotelomer alcohols, fluorotelomer acids, and short- and long-chain perfluorinated acids in water and biota. Journal of Chromatography, 2005, 1093 (A), 89–97

(21)

[15] EPA. SYNTHETIC PRECIPITATION LEACHING PROCEDURE.1994, found on:

https://www.epa.gov/hw-sw846/sw-846-test-method-1312-synthetic-precipitation-leaching-procedure opened on 18-04-2016

[16] Minhee, K. et al. Compositional Effects on Leaching of Stain-Guarded (Perfluoroalkyl and

Polyfluoroalkyl Substance-Treated) Carpet in Landfill Leachate. Environmental Science & Technology. 2015, 49 (11), 6564-6573

(22)

Appendix

Table A1: Information of relevant chemicals used during experiments

Chemical Grade Purity

(%)

Producer CAS (#)

Nitric acid For trace analysis ~65 Merck 7697-37-2

Sulphuric acid ACS reagent, reag.

ISO

95-97 Sigma-Aldrich

7664-93-9

Hydrochloric acid Trace metal analysis 36.5-38 J.T Baker 7647-01-0

Sodium hydroxide 1M 1.000N FLUKA 1310-73-2

Acetic acid ACS reagent, reag.

ISO

> 99.8 Sigma-Aldrich

64-19-7

Methanol Ultra HPLC Grade J.T.Baker 65-56-1

Potassium chloride FLUKA guarantee ≥ 99.5 FLUKA

Chemika

7447-40-7

Glycine Bio Ultra ≥ 99.0 Sigma Life 56-40-6

Sodium bicarbonate ≥ 99.5 FLUKA

analytical

144-55-8

Tris(hydroxymethyl)aminomethane Ultrapure Applichem 77-86-1

Ammonium formate Bio ultra ≥ 99.0 FLUKA

analytical

560-69-2

Ammonium acetate Reagent ≥ 98.0 Sigma

Aldrich

631-61-8

Ammonium hydroxide solution reag. ISO, reag. Ph. Eur. puriss. p.a.

Sigma Aldrich

30501-2.5L

Potassium phosphate monobasic ≥ 99.0 Sigma Life 7778-77-0

(23)

Table A2: Instrumental settings for PFAAs and FOSA analyses. Abbreviation MS/MS mass transition (m/z-> m/z) Fragmentor voltage (V) Collision energy (V) Ionization mode Isotope-labeled standard PFBA 213.0  169.0 60 3 Negative 13_C 4-PFBA PFPeA 263.0  219.0 60 3 Negative 13_C 5-PFPeA PFHxA 313.0  269.0 80 3 Negative 13_C 2-PFHxA PFHpA 363.1  319.0 80 4 Negative 13_C 4-PFHpA PFOA 413.0  369.0 80 4 Negative 13_C 4-PFOA PFNA 463.0  419.0 100 5 Negative 13_C 5-PFNA PFDA 513.0  468.9 100 5 Negative 13_C 2-PFDA PFUnDA 562.9  518.9 100 6 Negative 13_C 2-PFUnDA PFDoDA 613.0  568.9 100 7 Negative 13_C 2-PFDoDA PFTrDA 663.0  618.9 100 7 Negative 13_C 2-PFUnDA PFTeDA 712.9  668.9 120 4 Negative 13_C 2-PFDoDA PFBS 299.0  80.0 150 35 Negative 18_O 2-PFHxS PFHxS 399.0  80.0 200 48 Negative 18_O 2-PFHxS PFHpS 449.0  80.0 150 45 Negative 18_O 2-PFHxS PFOS 499.0  80.0 200 48 Negative 13_C 4-PFOS FOSA 498.1  78.0 200 35 Negative 13_C 8-FOSA 13_C 4-PFBA 217.0  172.0 60 3 Negative 13_C 5-PFPeA 268.0  222.9 60 3 Negative 13_C 2-PFHxA 315.0  270.0 80 3 Negative 13_C 4-PFHpA 367.0  321.9 80 4 Negative 13_C 4-PFOA 416.9  371.9 80 4 Negative 13_C 5-PFNA 468.0  423.0 100 5 Negative 13_C 2-PFDA 515.0  470.0 100 5 Negative 13_C 2-PFUnDA 565.0  520.0 100 6 Negative 13_C 2-PFDoDA 615.0  569.9 100 7 Negative 18_O 2-PFHxS 403.0  84 200 48 Negative 13_C 4-PFOS 503.0  80 200 48 Negative 13_C 8-FOSA 506.1  78 200 35 Negative

(24)

Table A3: Output for all PFASs PF BA PF Pe A PF H xA PF H pA PF O A PF N A PF D A PF U nD A PF D oD A PF Tr D A PF Te D A PF BS PF H xS PF H pS To t-PF O S L-PF O S Br -P FO S FO SA 4:2 F TS A 6:2 F TS A 8:2 F TS A 6:2 F TCA 8:2 F TCA 10 :2 F TCA Le ac hi ng p H 2 .0 *0 .2 2 0. 19 0. 36 *0 .0 3 0. 07 0 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 3 .5 0 0. 35 0. 67 0. 2 0. 23 0. 04 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 4 .2 0. 32 0. 25 0. 44 *0 .0 6 0. 04 0 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 4 .5 0 0. 25 0. 63 0. 2 0. 26 0. 06 *0 .0 2 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 5 .0 0 0. 2 0. 58 0. 16 0. 22 0. 04 *0 .0 1 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 7 .0 *0 .1 5 *0 .1 2 0. 46 0. 11 0. 1 *0 .0 06 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 8 .5 *0 .1 7 0. 22 0. 58 0. 14 0. 21 0. 03 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng p H 1 0 0. 23 0. 25 0. 67 0. 22 0. 27 0. 05 *0 .0 09 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 2 0° C *0 .2 0. 2 0. 3 *0 .0 3 *0 .0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Le ac hi ng 4 0° C *0 .2 0. 21 0. 48 *0 .0 9 0. 05 *0 .0 09 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Le ac hi ng 1 h ou r 0 0 *0 .0 3 0 *0 .0 2 0 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 4 h ou rs 0 *0 .0 3 0. 1 0. 03 0. 03 0. 00 4 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 1 d ay 0 *0 .0 8 0. 33 *0 .0 8 0. 12 *0 .0 2 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 2 d ay s 0 *0 .1 2 0. 35 0. 09 0. 13 0. 03 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 4 d ay s 0 0. 14 0. 41 0. 12 0. 17 0. 03 *0 .0 06 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 8 d ay s 0 0. 16 0. 43 0. 14 0. 23 0. 04 *0 .0 1 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 1 6 da ys 0 *0 .2 0. 53 0. 19 0. 38 0. 07 *0 .0 1 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng 2 1 da ys 0 *0 .2 1 0. 61 0. 22 0. 44 0. 06 *0 .0 2 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng R ai nw ate r 0. 24 0. 21 0. 41 0. 06 0. 02 0 0 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng Co lo r R eu s Bo nte W as 0. 23 0. 25 0. 68 0. 26 *0 .3 9 0. 18 0. 09 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 Le ac hi ng N ik w ax 0. 21 0. 31 0. 75 0. 26 0. 39 0. 14 0. 05 0 0 0 0 0 0 0 0 0 0 NA 0 0 0 0 0 0 M eth an ol E xtr ac ti on 1 0. 25 0. 3 0. 86 0. 27 0. 54 0. 24 0. 17 *0 .0 5 *0 .0 3 0 0 0 0 0 0 *0 .0 2 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on 2 *0 .1 6 0. 25 *0 .6 7 0. 24 0. 52 0. 24 0. 14 *0 .0 6 *0 .0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 1 h le ac hi ng 0 0. 18 0. 51 0. 16 0. 33 0. 14 0. 12 0. 03 0. 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 4 h le ac hi ng 0 0. 15 0. 48 0. 15 0. 32 0. 14 0. 13 0. 04 0. 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 1 d ay le ac hi ng 0 0. 11 0. 32 0. 11 0. 27 0. 12 0. 12 0. 03 0. 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 2 d ay s le ac hi ng 0 *0 .0 9 0. 28 0. 1 0. 23 0. 12 0. 11 *0 .0 2 0. 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 4 d ay s le ac hi ng 0 0. 07 0. 23 0. 09 0. 21 0. 11 0. 1 0. 02 0. 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 8 d ay s le ac hi ng 0 *0 .0 5 0. 19 0. 07 0. 19 0. 1 0. 11 0. 03 0. 02 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 1 6 da ys le ac hi ng 0 *0 .0 3 0. 1 0. 04 0. 15 0. 14 0. 13 0. 05 0. 02 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M eth an ol E xtr ac ti on a fte r 2 1 da ys le ac hi ng 0 <0 .0 3 *0 .0 6 *0 .0 3 0. 13 0. 07 0. 08 *0 .0 2 0. 02 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

(25)

Table A4: All recoveries for every sample Sa m p le 1 3 C 4 PF BA 1 3 C 5 PF Pe A 1 3 C 2 PF H xA 1 3 C 4 PF H p A 1 3 C 4 PF O A 1 3 C 5 PF N A 1 3 C 2 PF D A 1 3 C 2 PF U n D A 1 3 C 2 PF D o D A 1 8 O 2 PF H xS 1 3 C 4 PF O S 1 3 C 8 F O SA 1 3 C 2 6 :2 F T SA L e a c h in g p H 2 .0 147% 166% 167% 168% 171% 148% 123% 108% 86% 160% 118% 1% 378% L e a c h in g p H 3 .5 23% 142% 97% 112% 134% 115% 100% 97% 85% 133% 107% 0% 118% L e a c h in g p H 4 .2 159% 181% 158% 154% 170% 153% 137% 112% 90% 146% 122% 0% 298% L e a c h in g p H 4 .5 23% 171% 122% 135% 164% 146% 122% 128% 102% 163% 135% 1% 129% L e a c h in g p H 5 .0 16% 131% 88% 102% 130% 110% 103% 107% 91% 138% 116% 0% 100% L e a c h in g p H 7 .0 102% 108% 95% 114% 116% 115% 93% 86% 69% 111% 99% 0% 165% L e a c h in g p H 8 .5 167% 182% 158% 179% 180% 168% 147% 137% 113% 173% 153% 1% 346% L e a c h in g p H 1 0 316% 328% 296% 315% 285% 267% 245% 232% 178% 300% 237% 1% 479% L e a c h in g 2 0 °C 137% 132% 124% 133% 131% 127% 109% 92% 75% 124% 102% 1% 237% L e a c h in g 4 0 °C 107% 117% 104% 118% 113% 98% 78% 72% 69% 102% 74% 1% 192% L e a c h in g 1 h o u r 12% 119% 129% 125% 142% 138% 123% 125% 96% 160% 131% 1% 127% L e a c h in g 4 h o u rs 25% 333% 325% 337% 414% 400% 344% 354% 271% 450% 379% 3% 334% L e a c h in g 1 d a y 8% 54% 48% 50% 63% 61% 56% 64% 52% 70% 65% 1% 47% L e a c h in g 2 d a ys 9% 94% 94% 96% 115% 117% 108% 109% 86% 126% 108% 1% 81% L e a c h in g 4 d a ys 8% 98% 94% 101% 117% 113% 101% 107% 83% 125% 105% 0% 86% L e a c h in g 8 d a ys 8% 71% 94% 100% 127% 122% 108% 104% 94% 115% 109% 1% 96% L e a c h in g 1 6 d a ys 2% 33% 59% 55% 49% 57% 68% 68% 67% 93% 87% 1% 58% L e a c h in g 2 1 d a ys 3% 22% 44% 44% 30% 34% 37% 47% 48% 65% 54% 0% 32% L e a c h in g R a in w a te r 181% 208% 178% 173% 176% 160% 123% 107% 87% 179% 126% 1% 288% Bl a n k R a in w a te r 129% 133% 141% 137% 155% 144% 123% 101% 68% 157% 122% 1% 194% L e a c h in g C o lo r R e u s Bo n te W a s 119% 160% 142% 50% 6% 13% 23% 33% 48% 44% 14% 0% 14% Bl a n k C o lo r R e u s Bo n te W a s 185% 190% 179% 14% 6% 8% 15% 31% 79% 18% 9% 0% 13% L e a c h in g N ik w a x 88% 151% 143% 131% 142% 155% 128% 117% 106% 131% 119% 0% 282% Bl a n k N ik w a x 9% 116% 139% 126% 121% 134% 132% 122% 127% 124% 119% 1% 217% M e th a n o l Ex tra c tio n 1 66% 104% 95% 98% 82% 84% 59% 37% 23% 67% 36% 12% 187% M e th a n o l Ex tra c tio n 2 77% 101% 91% 93% 77% 78% 54% 34% 20% 63% 34% 16% 180% M e th a n o l Ex tra c tio n a ft e r 1 h le a c h in g 37% 152% 157% 195% 213% 150% 99% 104% 73% 140% 123% 27% 226% M e th a n o l Ex tra c tio n a ft e r 4 h le a c h in g 31% 135% 153% 180% 196% 147% 91% 94% 69% 132% 113% 31% 231% M e th a n o l Ex tra c tio n a ft e r 1 d a y le a c h in g 22% 132% 162% 180% 194% 142% 87% 94% 75% 129% 111% 34% 237% M e th a n o l Ex tra c tio n a ft e r 2 d a ys le a c h in g 21% 121% 141% 159% 188% 149% 85% 85% 65% 129% 114% 33% 252% M e th a n o l Ex tra c tio n a ft e r 4 d a ys le a c h in g 27% 142% 182% 212% 234% 175% 102% 103% 74% 161% 136% 23% 295% M e th a n o l Ex tra c tio n a ft e r 8 d a ys le a c h in g 19% 126% 144% 169% 203% 144% 95% 95% 82% 127% 110% 42% 245% M e th a n o l Ex tra c tio n a ft e r 1 6 d a ys le a c h in g 14% 119% 160% 197% 218% 154% 99% 101% 89% 148% 123% 32% 275% M e th a n o l Ex tra c tio n a ft e r 2 1 d a ys le a c h in g 8% 67% 77% 116% 95% 80% 46% 48% 39% 86% 75% 14% 144%