Background/Objectives
Environmental contamination with per- and polyfluoroalkyl substances (PFAS) has received increasing interest in the early 21st century, since identification of PFAS in human blood and animal tissues globally. The use of PFAS in aqueous film forming foams (AFFF) for training and fighting hydrocarbon fires has received particular attention, as this has proved an important route to the environment in Australia and abroad.

The Australian Department of Defence (Defence) is undertaking a nationwide program to investigate and manage PFAS contamination from legacy AFFF use on Defence sites. Investigations have been undertaken at 28 sites across the country, using an accelerated tiered process, responsive to community and regulator expectations, to rapidly assess the extent and associated risks from contamination by PFAS compounds. Investigations focused on migration and exposure pathways, and as a result have collected information about soils, waters and biota under different environmental conditions.

When planning for redevelopment works in an area that may be contaminated by legacy use of AFFF it is usually necessary to classify soils for management, reuse, disposal or destruction. Sampling plans should have regard to the conceptual site model, the properties of the contaminant, the size and geometry of the nominated hotspot and the required degree of confidence.

This presentation will review available soil data from multiple Defence site investigations. Spatial relationships and correlations between surface soil PFAS concentrations and other soil parameters will be presented, with a view to guiding future sampling plans.

Approach/Activities
Over 15,000 soil samples from the Defence investigations will be classified into a range of categories based on soil type, depth, climate type, land use and mechanism of contamination. Samples were collected by qualified environmental consultants using techniques aiming to avoid PFAS contamination. Analyses were performed using liquid chromatography tandem mass spectrometry by accredited commercial laboratories. Using R and QGIS, a range of geostatistical tools will be applied to the data to determine spatial relationships and correlate these with the soil categories for individual PFAS. A key approach will be to develop an empirical variogram for each situation and then develop a correlation matrix based on the nugget, range and sill of each variogram.

Results/Lessons Learned
It is expected that this analysis will illustrate the degree to which PFAS concentrations in soils are spatially correlated with factors such as climate, depth, soil type and mechanism of contamination. Understanding this variability in the context of the conceptual site model will facilitate the development of more efficient sampling plans that provide a better quantification of uncertainty.

Andrew Mitchell, Executive Project Director, RPS Group
Andrew Mitchell currently leads a technical policy team in supporting the Australian Department of Defence in its investigation and management of PFAS contamination arising from legacy use of AFFF. He has a Bachelor of Environmental Engineering, a Master of Public Administration and over 20 years’ experience in solving environmental problems in regulation, government and consulting.

Background/Objectives
Multiple site investigations where assessment of per- and polyfluoroalkyl substances (PFAS) at fire training areas (FTA) using advanced characterisation tools, such as the total oxidisable precursor (TOP) assay, have determined that high concentrations of PFAS, including precursors are present in surficial soil and vadose zone. The results from several site investigations of FTAs using advanced analytical tools are compared and combined with a detailed understanding of the partitioning behaviour of fluorosurfactants to explain why unsaturated zone soils comprise a reservoir of PFAS.

Approach/Activities
Groundwater and soil samples were characterized for PFAS at multiple sites using advanced analytical tools. A detailed literature review of publications considering the behaviour of fluorosuractants was performed to shed light on their partitioning behaviour.

Results/Lessons Learned
The vertical and horizontal delineation of PFAS at this site will be presented in relation to the site’s hydrogeology and lithology. The self-assembly behaviour of fluorosurfactants is characterized by a strong tendency to form vesicles and lamellar phases rather than micelles, primarily due to the bulkiness of the perfluoroalkyl chain, that tends to decrease the curvature of the aggregates they form in solution.​ These bulky supramolecular structures can further assemble to create a liquid crystalline phase which can grow to micron sized structures. These reservoirs of PFAS, bound to soils, primarily adhere to surfaces, so deposits such as clays, which comprise those with greatest surface area are primary sites for self-assembly and growth of supramolecular PFAS reservoirs, which form following repeat fire training.

The often frequent and repeat, historical applications of class B firefighting foams such as aqueous film forming foams (AFFF) to open ground are considered to be responsible for this mass distribution, where amphiphilic PFAS may primarily be stored in the vadose zone. When assessing the distribution of PFAS in soil and groundwater as a result of repeat fire training or equipment testing, the fate and behavior of all of the components of AFFF, such as glycols, hydrocarbon surfactants need to be considered, in addition to the PFAS, to understand fate and transport of PFAS originating from AFFF.

Self-assembly of C8 PFAS into contiguous bilayers has been reported to occur at concentrations as low as 50 mg/L, meaning that the g/L concentrations encountered in AFFF as discharged to ground, is far greater than that needed for growth of these lamellar, vesicles and microtubules on soil surfaces. The glycol components of foams are solvents which allow high concentrations of fluorosurfactants to dissolve. When AFFF is deposited to ground, the glycols will rapidly biodegrade meaning their solvent properties are lost, which increases the propensity of PFAS to partition to self-assembled supramolecular phases. A model will be described which details the interaction of PFAS in AFFF with soils, explaining why FTAs remain as continuing source of PFAS to groundwater plumes for many decades. The implications of self-assembly of amphiphilic PFAS at air:solid and water:solid interfaces as supramolecular structures, will be discussed considering development of conceptual site models and remediation.

Ian Ross, Senior Technical Director, Arcadis UK
Ian Ross, PhD, is a senior technical director with Arcadis and the company’s global PFAS lead. He has focused on remediation for more than 25 years, designing and implementing innovative chemical, physical, and biological remediation technologies. He is currently focused on risk management and remediation of PFAS, with more than 14 years of experience working on PFAS projects.

Per- and polyfluoroalkyl substances (PFAS) encompass a wide range of chemicals with highly disparate properties and behaviour in the environment. Regulatory criteria or guidelines have been established in various jurisdictions for a subset of PFAS, typically selected perfluoroalkyl acids (PFAA), such as perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA). Unlike chemicals that frequently drive sediment remediation, e.g., mercury and polychlorinated biphenyls (PCBs), PFAA are expected to depurate relatively rapidly from sediments following source control (due to low partitioning to sediments) and not constitute a long-term sediment concern. However, the presence of certain PFAS (“precursors”), which transform to PFAA and may be more tightly bound to sediments, indicates a need for modeling to better understand chemical longevity in the sediments.

The objective of this study is to provide insights to whether there exist certain conditions under which PFAS may constitute a sediment management concern, including due to the formation of PFAA from precursors. Precursors with low mobility in sediments (e.g., those with high partitioning) could serve as a long-term source of PFAA, but only if the transformation rate is high enough to generate appreciable quantities of the PFAA end-products. This establishes a fate and transport trade-off, because chemicals that partition strongly to the sediments are often not bioavailable enough to transform rapidly.

A chemical fate and transport model representing the sediment bed and overlying water column was used to simulate potential scenarios to evaluate how combinations of precursor transformation rates, partitioning, diffusion, groundwater flux, and advective transport affect PFAA concentrations. Evaluations focused on the recovery rate of sediments following cessation of the PFAS source (e.g., a groundwater plume discharging through the sediment bed to surface water).

Modeling results indicate that precursors with high partition coefficients can result in an extended duration of PFAA presence, but that the high partition coefficient limits bioavailability for precursor transformation and therefore only low concentrations of PFAA accumulate in porewater and surface water. These low concentrations may not accumulate to ecologically relevant levels in fish. Canada considers PFOS to be bioaccumulative in fish, but it is less bioaccumulative than PCBs and mercury. Furthermore, most jurisdictions do not consider short-chain PFAA to be bioaccumulative in fish. Bioaccumulation is relatively low because depuration across the gills is relatively high. Still, regulations describing acceptable levels of PFOS are evolving, with some jurisdictions (e.g., Australia) having much lower acceptable levels for surface water than Canada. Canada does not have Federal Environmental Quality Guidelines for PFOS in sediment or porewater. Should these be promulgated, residual levels of PFOS in sediment and porewater may also be of regulatory concern.

This analysis also illustrates how ongoing remedial actions, e.g., for legacy contaminants, may, or may not, be addressing PFAS. For example, sediment capping designed to limit migration of highly hydrophobic chemicals may not be effective for the more mobile precursors and most, if not all, PFAA. Source control of legacy contaminants, followed by dredging of residuals, will not be effective for PFAS unless the PFAS source is also controlled. Under most conditions, control of the PFAS source will lead to significant reductions in PFAA concentrations, potentially obviating the need for sediment remediation. Once PFAA exposures are reduced, the high depuration rate of PFAAs will allow fish concentrations to drop rapidly.

Daniel Opdyke, Senior Managing Engineer, Anchor QEA, LLC
Daniel Opdyke is a technical expert in multiple surface water and groundwater subjects, including design of monitoring programs; verification, analysis, and visualization of data; development of conceptual models; design and calibration of statistical and mechanistic models; and documentation and presentation of results to stakeholders. He has applied these skills to a variety of projects and disciplines, including groundwater availability, groundwater contaminant fate and transport, surface water availability, negotiation of water rights, surface water eutrophication, surface water contaminant fate and transport, groundwater and surface water interactions, chemical bioaccumulation, environmental flows, and freshwater inflows to estuaries. More recently, Daniel has worked in the area of PFAS, understanding their fate, transport, and potential bioaccumulation in surface and groundwater systems.