The increasing awareness towards PFAS pollution makes it increasingly important to treat. Many agencies over the globe, including the Swedish Food Agency, are introducing ever so strict control over PFASs in drinking water. One of Norrvatten’s back-up drinking water plants, the Hammarby plant, is a groundwater plant that saw elevated concentrations of PFASs (sum of 32 PFASs~80 ng/L) and total uranium (~109 µg/L). The aim of this work is to investigate the possible remediation for PFAS and uranium pollution in the groundwater. The main objective is to examine the possibility of SorbixTM, a strong-base anion exchange resins provided by ETC2, adsorbing PFAS and uraniumfrom the groundwater. Furthermore, the optimal solution of groundwater and sodium chloride that prevents microbial growth is to be investigated. This is of interest considering the intermittent operation of the Hammarby plant, where water would be allowed to stagnant for longer periods thus being prone to microbial growth. The effect of microorganisms on the resins is of concern. Biogrowth experiments were conducted using four different brines: 0%, 5%, 15%, and 25%. Identical masses of AIX resins were soaked in these solutions for 19-29 days in order to investigate whether the brine is effective at inhibiting biogrowth. The tests were analyzed mainly using plate count method, and secondly using flow cytometry. For uranium and PFAS removal, batch experiments using (1) resins that were previously soaked in brine were performed, (2) resins that were exposed to microorganisms as well as (3) new unused resins. Results showed that brines with 15% salinity or higher are capable of completely stopping microbial growth. The SorbixTM was capable of removing PFAS to below detection levels, as well as 99% removal of total uranium. To further develop on this study, column experiments with continuous water flow should be investigated. Continuous flow is a more accurate representation of the real case and will give knowledge on additional information regarding the breakthrough volume of the individual PFASs, the required water flow relative to available resin area that ensures acceptable removals, and how the competitiveness of uranium affects PFAS adsorption. It is also recommended to investigate the potential for regenerating the AIX resins. The Swedish Food Agency restricts the chemicals that are allowed to come in contact with any drinking water equipment. This means that some of the regenerative substances used in other studies are prohibited in Sweden. / Den växande uppmärksamheten kring PFAS gör det allt viktigare att avlägsna dessa från dricksvatten. Flera myndigheter över hela världen, inklusive Livsmedelsverket, inför gränsvärden och kontroller på PFAS i dricksvatten som blir allt striktare. En av Norrvattens reservvattenverk, Hammarby, är en grundvattenanläggning som visade förhöjda koncentrationer av PFAS och uran, där summan av de 32 mest förekommande PFASs är 80 ng/L och den totala uranhalten är 109 µg/L. Syftet med detta arbete är att undersöka möjligheten till PFAS och uran borttagning från grundvattnet. Huvudmålet är att undersöka möjligheten för SorbixTM, en starkbasisk anjonbytesharts som är framtagen av ETC2 Sverige (ETC2, n.d.), att adsorbera PFAS och uran från grundvattnet. Dessutom ska den optimala lösningen av grundvatten och natriumklorid, som förhindrar mikrobiell tillväxt, undersökas. Detta är av intresse med tanke på den intermittenta driften av Hammarby-anläggningen, där vattnet i ledningarna är stillastående under längre perioder och därmed blir utsatt för mikrobiell tillväxt. Effekten av mikroorganismer på jonbytesharts är, enligt andrastudier, bekymmersam. Tillväxtexperimenten utfördes genom användning av fyra olika saltlösningar: 0%, 5%, 15%och 25%. Identiska massor av anjonbytesharts blötlades i dessa lösningar i 19-29 dagar för att undersöka biologisk tillväxt. Testerna analyserades huvudsakligen med tre-dagars-tillväxt och imindre utsträckning med flödescytometri.För uran- och PFAS-avlägsnande så användes standardiserad skaktester med (1) jonbytesharts somtidigare blötlagts i saltvatten, (2) jonbytesharts som exponerats för mikroorganismer samt (3) nytt och oanvänt jonbytesharts. Resultatet visade att lösningar med 15% salthalt eller högre kan bromsa den mikrobiella tillväxten helt. SorbixTM lyckades ta bort PFAS till under detektionsnivåer, samt avlägsnade över 99% av den uppmätta uranen i grundvattnet. För vidare studier så rekommenderas kolonnförsök med kontinuerligt vattenflöde. Kontinuerligt flöde är en mer exakt representation av det verkliga fallet och kommer att ge ytterligare information såsom: breakthrough volymen för de enskilda PFASs, optimala vattenflödet i förhållande till tillgänglig yta av jonbytesharts som säkerställer acceptabel avlägsnande samt hur stark konkurrensen mellan uran och PFAS blir och vilken av de som adsorberas/desorberas först. Dessutom rekommenderas undersökningar på regenereringen av jonbytesharts, med tanke till de begränsade regenereringsämnen som är tillåtna enligt Livsmedelsverket. Många regenerativa ämnensom används i andra studier är förbjudna att använda i Sverige.

The phytotoxicities of monensin, the 10:2 saturated fluorotelomer carboxylic acid (10:2 FTCA), and atrazine to freshwater macrophytes were investigated in three microcosm-based assessments. Both responses of plants grown as individuals in “cone-tainers” and those grown in mixed and monocultures were examined to permit comparison of the toxicological sensitivities of macrophytes under each planting design, and investigation of whether interactions between neighbours may modify plant response to an environmental contaminant. Exposures of monensin and the 10:2 FTCA at environmentally relevant concentrations were found to produce few significant effects in the higher aquatic plants across all growth conditions, thus direct comparisons of effective concentrations were not conducted. Significant differences between relative growth rates (RGR) of plant grown in assemblages versus individually indirectly indicate that over longer exposure durations toxicity may be underestimated using the individual “cone-tainer” method. RGRs and sensitivities of plants to atrazine were found to be in the same range across planting methods, demonstrating that responses of aquatic plants in the individual-test system reflected those observed in model populations and two-species communities. A lack of observed relations between plants in the mixed and monoculture tests, however, meant that the potential for modification of toxicity through plant interactions was not investigated.

The phytotoxicities of monensin, the 10:2 saturated fluorotelomer carboxylic acid (10:2 FTCA), and atrazine to freshwater macrophytes were investigated in three microcosm-based assessments. Both responses of plants grown as individuals in “cone-tainers” and those grown in mixed and monocultures were examined to permit comparison of the toxicological sensitivities of macrophytes under each planting design, and investigation of whether interactions between neighbours may modify plant response to an environmental contaminant. Exposures of monensin and the 10:2 FTCA at environmentally relevant concentrations were found to produce few significant effects in the higher aquatic plants across all growth conditions, thus direct comparisons of effective concentrations were not conducted. Significant differences between relative growth rates (RGR) of plant grown in assemblages versus individually indirectly indicate that over longer exposure durations toxicity may be underestimated using the individual “cone-tainer” method. RGRs and sensitivities of plants to atrazine were found to be in the same range across planting methods, demonstrating that responses of aquatic plants in the individual-test system reflected those observed in model populations and two-species communities. A lack of observed relations between plants in the mixed and monoculture tests, however, meant that the potential for modification of toxicity through plant interactions was not investigated.

Perfluorinated carboxylates (PFCAs) are persistent and ubiquitous in the environment. Humans are exposed to PFCAs through direct and indirect sources, although the relative importance of each is uncertain. Direct sources of PFCAs have been attributed to two primary fluorochemical manufacturing processes: electrochemical fluorination (ECF) and telomerization. A focus of this thesis was to elucidate an additional direct source of PFCAs resulting from the direct fluorination of polyolefin materials. High density polyethylene bottles with varying levels of fluorination were observed to contain significant amounts of PFCAs, particularly those with carbon chain-lengths ≤ C6, marking an unexplored source of PFCA exposure. PFCAs are also produced indirectly from the biotransformation of fluorotelomer-based compounds, such as polyfluoroalkyl phosphate esters (PAPs) and fluorotelomer alcohols (FTOHs). During this transformation process, two predominant classes of metabolic intermediates are formed: the fluorotelomer unsaturated aldehydes (FTUALs) and the fluorotelomer unsaturated carboxylic acids (FTUCAs). Another focus of this thesis was to examine the reactivity of FTUALs and FTUCAs with endogenous nucleophiles such as glutathione (GSH), select amino acids, and model proteins. FTUALs formed adducts with all nucleophiles examined, where those having shorter carbon chain lengths (i.e. 6:2 and 8:2 FTUAL) were more reactive than longer carbon chains (i.e. 10:2 FTUAL). By contrast, FTUCAs had comparably limited reactivity; although FTUCAs showed mild reactivity with GSH, they did not react with any other nucleophiles. In vitro and in vivo experiments were carried out to determine the extent of protein binding formed from the biotransformation of fluorotelomer-based compounds, including the 8:2 FTOH and the 6:2 PAP diester. A significant portion of these biotransformations yielded covalent protein binding at nmol/mg protein concentrations. Protein adducts were observed predominantly in rat liver and also in plasma and kidney. The formation of reactive intermediates may be toxicologically important through protein deactivation. Cellular toxicity of FTUALs was significantly higher compared to PFCAs and the acid metabolic intermediates (i.e. FTUCAs). The EC50 values calculated from dose-response incubations were dependant on chain length and functional group. The work in this thesis examined an unexplored consequence of indirect exposure to PFCAs, potentially impacting the relative importance of PFCA exposure sources.

Perfluorinated carboxylates (PFCAs) are persistent and ubiquitous in the environment. Humans are exposed to PFCAs through direct and indirect sources, although the relative importance of each is uncertain. Direct sources of PFCAs have been attributed to two primary fluorochemical manufacturing processes: electrochemical fluorination (ECF) and telomerization. A focus of this thesis was to elucidate an additional direct source of PFCAs resulting from the direct fluorination of polyolefin materials. High density polyethylene bottles with varying levels of fluorination were observed to contain significant amounts of PFCAs, particularly those with carbon chain-lengths ≤ C6, marking an unexplored source of PFCA exposure. PFCAs are also produced indirectly from the biotransformation of fluorotelomer-based compounds, such as polyfluoroalkyl phosphate esters (PAPs) and fluorotelomer alcohols (FTOHs). During this transformation process, two predominant classes of metabolic intermediates are formed: the fluorotelomer unsaturated aldehydes (FTUALs) and the fluorotelomer unsaturated carboxylic acids (FTUCAs). Another focus of this thesis was to examine the reactivity of FTUALs and FTUCAs with endogenous nucleophiles such as glutathione (GSH), select amino acids, and model proteins. FTUALs formed adducts with all nucleophiles examined, where those having shorter carbon chain lengths (i.e. 6:2 and 8:2 FTUAL) were more reactive than longer carbon chains (i.e. 10:2 FTUAL). By contrast, FTUCAs had comparably limited reactivity; although FTUCAs showed mild reactivity with GSH, they did not react with any other nucleophiles. In vitro and in vivo experiments were carried out to determine the extent of protein binding formed from the biotransformation of fluorotelomer-based compounds, including the 8:2 FTOH and the 6:2 PAP diester. A significant portion of these biotransformations yielded covalent protein binding at nmol/mg protein concentrations. Protein adducts were observed predominantly in rat liver and also in plasma and kidney. The formation of reactive intermediates may be toxicologically important through protein deactivation. Cellular toxicity of FTUALs was significantly higher compared to PFCAs and the acid metabolic intermediates (i.e. FTUCAs). The EC50 values calculated from dose-response incubations were dependant on chain length and functional group. The work in this thesis examined an unexplored consequence of indirect exposure to PFCAs, potentially impacting the relative importance of PFCA exposure sources.

Per- and polyfluorinated substances (PFAS) are used in the production of ski wax to reduce the friction between the snow and the ski. In this occupational study of ski wax technicians’ exposure to PFAS and particulate aerosol we have collected whole blood (wb) (n =94), air (n =84) and aerosol (n =159) samples at World Cup events from 2007-2011. We have analysed the blood, air and aerosol with respect to 13 perfluorocarboxylic acids (PFCAs), 4 perfluorosulfonic acids (PFSAs), 3 fluorotelomer alcohols (FTOHs), 3 fluorotelomer acids (FTCAs) and 3 unsaturated fluorotelomer acids (FTUCAs). Further, we assessed the exposure to 3 particulate aerosol fractions (inhalable, respirable and total aerosol) in air. In comparison to a general population, several of the PFCA blood levels are elevated in the technicians’, primarily  erfluorooctanoate (PFOA) and perfluorononate (PFNA) with concentrations up to 628 and 163 ng/mL wb, respectively. Further,  we detected FTUCAs and FTCAs in the blood, suggesting biotransformation of FTOHs to PFCAs. The metabolites 5:3 and 7:3 FTCA were detected in all blood samples at levels up to 6.1 and 3.9 ng/mL wb. Levels of perfluorohexadecanoic acid PFHxDA) and perfluorooctadecanoic acid (PFOcDA) were detected in the technician’s blood at mean concentration up to 4.22 ng/mL wb and 4.25 ng/mL wb. The FTOH levels in air of the wax cabin during work ranged up to 997 000 ng/m3 (average=114 000 ng/m3 ) and PFOA up to 4 890 ng/m3 (average= 526 ng/m3 . FTOHs were not detected in aerosols but PFOA showed average levels of 12 000 ng/m3 (range=1 230- 46 900 ng/m3 ). The occupational exposure limit (OEL) of 2 mg/m3 was exceeded in 37% of the personal measurements with aerosol  concentrations up to 15 mg/m3 . Keywords : Perfluorinated, polyfluorinated, FIS, occupational exposure, ski wax,  iotransformation, metabolism, fluorotelomer alcohol, fluorotelomer acid, aerosol, dust, UPLC/MS-MS, GC/MS-MS

京都大学 / 0048 / 新制・課程博士 / 博士(地球環境学) / 甲第18628号 / 地環博第123号 / 新制||地環||25(附属図書館) / 31528 / 京都大学大学院地球環境学舎環境マネジメント専攻 / (主査)教授 藤井 滋穂, 教授 梶井 克純, 准教授 田中 周平 / 学位規則第4条第1項該当 / Doctor of Global Environmental Studies / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第20814号 / 工博第4418号 / 新制||工||1686(附属図書館) / 京都大学大学院工学研究科都市環境工学専攻 / (主査)教授 伊藤 禎彦, 教授 田中 宏明, 准教授 西村 文武 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Fluorotelomer alcohols (FTOHs, F(CF2)nCH2CH2OH) are emerging contaminants in the environment. Biodegradation of 6:2 and 8:2 FTOHs has been intensively studied using soils and activated sludge. However, little is known about the bacteria responsible for biotransformation of FTOHs. This study deciphered factors affecting biodefluorination of FTOHs and their metabolites, and developed three effective FTOH-degrading consortia. Two alkane-degrading Pseudomonas strains (P. oleovorans and P. butanovora) can defluorinate 4:2, 6:2 and 8:2 FTOHs, with a higher degree of defluorination for 4:2 FTOH. According to the identified metabolites, P. oleovorans transformed FTOHs via two pathways I and II. Pathway I led to formation of x:2 ketone (x = n-1), x:2 sFTOH and perfluorinated carboxylic acids (PFCAs). Pathway II resulted in the formation of x:3 polyfluorinated acid and relatively minor shorter-chain PFCAs. Conversely, P. butanovora transformed FTOHs by pathway I only. Mycobacterium vaccae JOB5 (a C1-C22alkane-degrading bacterium) and P. fluorescens DSM 8341 (a fluoroacetate-degrading bacterium) can transform 6:2 FTOH via both pathways I and II with the formation of odd-numbered short-chain PFCAs. In the presence of dicyclopropylketone or formate, P. oleovorans transformed 6:2 FTOH six times faster and produced odd-numbered PFCAs. P. butanovora, utilized both pathways I and II in the presence of lactate, and it also produced odd-numbered PFCAs. Unlike P. oleovorans, P. fluorescens DSM 8341 could slightly convert 5:3 polyfluorinated acid (a key metabolite during 6:2 FTOH degradation, [F(CF2)5CH2CH2COOH]) to 4:3 acid and PFPeA via one-carbon removal pathways. Three FTOH-degrading consortia transformed FTOHs, with enhanced removal of FTOHs in the presence of n-octane. A higher copy number of alkB gene was found to correspond to better removal of FTOHs, suggesting that alkane-degrading bacteria might be the key degraders in the enrichments. The three enrichment cultures showed a similar microbial community structure. This is the first study reporting that pure strains of alkane- and fluoroacetate-degrading bacteria can bio-transform FTOHs via different or preferred transformation pathways to remove multiple –CF2– groups from FTOHs to form shorter-chain PFCAs, and to other perfluorinated acids. The results of this study also suggest that enhanced FTOH biodegradation is possible through co-substrate addition and/or using enrichment cultures.