Occurrence of Per- and Polyfluoroalkyl Substances (PFAS) in Source Water and Their Treatment in Drinking Water.

Per-and polyfluoroalkyl substances (PFAS) occurrence in drinking water and treatment methods for their removal are reviewed. PFAS are fluorinated substances whose unique properties make them effective surface-active agents with uses ranging from stain repellants to fire-fighting foams. In response to concerns about drinking water contamination and health risks from PFAS exposure, the United States Environmental Protection Agency published Health Advisories (HAs) for perfluorooctanoic acid and perfluorooctane sulfonic acid. The occurrence of six PFAS in drinking water has been reported in the Third Unregulated Contaminant Monitoring Rule (UCMR3), and subsequent analysis of the dataset suggested that four percent of water systems reported at least one detectable PFAS compound and 1.3 percent of water systems reported results above the HAs. Many treatment technologies have been evaluated in the literature, with the most promising and readily applied treatment technologies being activated carbon, anion exchange resins, and high-pressure membrane systems. From these data and literature reports, research and data gaps were identified and suggestions for future research are provided.

Role of grinding method on granular activated carbon characteristics.

A coconut shell (AC1230CX) and a bituminous coal based (F400) granular activated carbon (GAC) were ground with mortar and pestle (MP), a blender, and a bench-scale ball milling unit (BMU). Blender was the most time-efficient for particle size reduction. Four size fractions ranging from 20 × 40 to 200 × 325 were characterized along with the bulk GACs. Compared to bulk GACs, F400 blender and BMU 20 × 40 fractions decreased in specific surface area (SSA, -23% and -31%, respectively) while smaller variations (-14% to 5%) occurred randomly for AC1230CX ground fractions. For F400, the blender and BMU size fraction dependencies were attributed to the combination of (i) radial trends in the F400 particle properties and (ii) importance of shear (outer layer removal) versus shock (particle fracturing) size reduction mechanisms. Compared to bulk GACs, surface oxygen content (At%-O1s) increased up to 34% for the F400 blender and BMU 20 × 40 fractions, whereas all AC1230CX ground fractions, except for the blender 100 × 200 and BMU 60 × 100 and 100 × 200 fractions, showed 25-29% consistent increases. The At%-O1s gain was attributed to (i) radial trends in F400 properties and (ii) oxidization during grinding, both of which supported the shear mechanism of mechanical grinding. Relatively small to insignificant changes in point of zero charge (pHPZC) and crystalline structure showed similar trends with the changes in SSA and At%-O1s. The study findings provide guidance for informed selection of grinding methods based on GAC type and target particle sizes to improve the representativeness of adsorption studies conducted with ground GAC, such as rapid small-scale column tests. When GACs have radial trends in their properties and when the target size fraction only includes larger particle sizes, manual grinding is recommended.

Anion Exchange Resin and Inorganic Anion Parameter Determination for Model Validation and Evaluation of Unintended Consequences during PFAS Treatment.

When implementing anion exchange (AEX) for per- and polyfluoroalkyl substances treatment, temporal drinking water quality changes from concurrent inorganic anion (IA) removal can create unintended consequences (e.g., corrosion control impacts). To understand potential effects, four drinking water-relevant IAs (bicarbonate, chloride, sulfate, and nitrate) and three gel-type, strong-base AEX resins were evaluated. Batch binary isotherm experiments provided estimates of IA selectivity with respect to chloride ( K x ∕ C ) for IA/resin combinations where bicarbonate < sulfate ≤ nitrate at studied conditions. A multi-IA batch experiment demonstrated that binary isotherm-determined K x ∕ C values predicted competitive behavior. Subsequent column experiments with and without natural organic matter (NOM) allowed for the validation of a new ion exchange column model (IEX-CM; https://github.com/USEPA/Water_Treatment_Models). IA breakthrough was well-simulated using binary isotherm-determined K x ∕ C values and was minimally impacted by NOM. Initial AEX effluent water quality changes with corrosion implications included increased chloride and decreased sulfate and bicarbonate concentrations, resulting in elevated chloride-to-sulfate mass ratios (CSMRs) and Larson ratios (LRs) and depressed pH until the complete breakthrough of the relevant IA(s). IEX-CM utility was further illustrated by simulating the treatment of low-IA source water and a change in the source water to understand the resulting duration of changes in IAs and water quality parameters.

Strong Base Anion Exchange Selectivity of Nine Perfluoroalkyl Chemicals Relevant to Drinking Water.

Selectivity with respect to chloride (KPFAS∕C) was determined for nine drinking water relevant perfluoroalkyl and polyfluoroalkyl substances (PFAS): perfluoro-2-propoxypropanoic acid (GenX), five perfluoroalkyl carboxylic acids (PFCAs), and three perfluoroalkyl sulfonic acids (PFSAs). Three single-use strong base anion exchange gel resins were investigated, targeting drinking water relevant equilibrium PFAS liquid concentrations (≤500 ng/L). Except for the longest carbon chain PFCA (perfluorodecanoic acid) and PFSA (perfluorooctanesulfonic acid) studied, PFAS followed traditional ion exchange theory (law of mass action), including increasing equilibrium PFAS liquid concentrations with increasing equilibrium chloride liquid concentrations. Overall, KPFAS∕C values were (i) similar among resins for a given PFAS, (ii) 1-5 orders of magnitude greater than the selectivity of inorganic anions (e.g., nitrate) previously studied, (iii) 2 orders of magnitude greater for the same carbon chain length PFSA versus PFCA, (iv) found to proportionally increase with carbon chain length for both PFSAs and PFCAs, and (v) similar for GenX and perfluorohexanoic acid (six-carbon PFCA). A multisolute competition experiment demonstrated binary isotherm-determined KPFAS∕C values could be applied to simulate a multisolute system, extending work previously done with only inorganic anions to PFAS. Ultimately, estimated KPFAS∕C values allow future extension and validation of an open-source anion exchange column model to PFAS.

Practical implications of perfluoroalkyl substances adsorption on bottle materials: Isotherms.

To assess the practical implications of various bottle materials used in anion exchange (IX) or granular activated carbon (GAC) isotherm experiments, adsorption of seven per- and polyfluoroalkyl substances (PFAS) onto three common bottle materials (silanized glass, polypropylene, and high-density polyethylene [HDPE]) were screened. Results were similar between bottle materials; therefore, only HDPE was used in a detailed bottle material isotherm study with 11 PFAS. For each PFAS, an HDPE bottle isotherm was generated with equilibrium liquid phase concentrations relevant to drinking water (<2000 ng/L). Percent PFAS recoveries between 90% and 103%, 85% and 114%, and 54% and 108% were determined for perfluoro-2-propoxypropanoic acid (GenX), five perfluoroalkyl carboxylic acids, and five perfluoroalkyl sulfonic acids (PFSA), respectively. These results indicated only the five PFSA adsorbed to the HDPE bottles in a concentration-dependent manner. Furthermore, linear isomer versions of two PFSA exhibited greater adsorption. For each PFSA studied, a linear isotherm was generated and used to develop guidance for conducting future IX and GAC isotherm studies. Specifically, the minimum initial isotherm concentration was established such that a maximum 1% loss would be expected to the HDPE bottles, resulting in required initial concentrations of the five PFSA between 21 and 75 times that of the design isotherm liquid equilibrium concentration.

Laboratory evaluation of oil spill bioremediation products in salt and freshwater systems.

Ten oil spill bioremediation products were tested in the laboratory for their ability to enhance biodegradation of weathered Alaskan North Slope crude oil in both freshwater and saltwater media. The products included nutrients to stimulate inoculated microorganisms, nutrients plus an oil-degrading inoculum, nutrients plus compounds intended to stimulate oil-degrading activity, or other compounds intended to enhance microbial activity. The product tests were undertaken to evaluate significant modifications in the existing official United States Environmental Protection Agency (EPA) protocol used for qualifying commercial bioremediation agents for use in oil spills. The EPA protocol was modified to include defined formulas for the exposure waters (freshwater, saltwater), a positive control using a known inoculum and nutrients, two negative controls (one sterile, the other inoculated but nutrient-limited), and simplified oil chemical analysis. Three analysts conducted the product test independently in each type of exposure water in round-robin fashion. Statistical tests were performed on analyst variability, reproducibility, and repeatability, and the performance of the various products was quantified in both exposure media. Analysis of variance showed that the analyst error at each time-point was highly significant (P values ranged from 0.0001 to 0.008, depending on water type and oil fraction). In the saltwater tests, six products demonstrated various degrees of biodegradative activity against the alkane fraction of the crude oil and three degraded the aromatic hydrocarbons by >10%. In the freshwater tests, eight products caused >20% loss of alkane hydrocarbons, of which five degraded the alkanes by >50%. Only four products were able to degrade polycyclic aromatic hydrocarbons (PAHs) by >20%, one of which caused 88% removal. However, when the variability of the analysts was taken into consideration, only one of the ten products was found to yield significant percent removals of the PAH fraction and only in freshwater. Viable microorganism population analysis (most-probable-number method) was also performed on every sample by each operator to measure the changes in aromatic and alkane hydrocarbon-degrading organism numbers. In general, little evidence of significant growth of either alkane- or PAH-degraders occurred among any of the ten products in either the saltwater or freshwater testing.