Formatted Title
The Influence of Tension-Driven Flow on the Transport of AFFF in Unsaturated Zone
Background/Objectives
The widespread detection of per- and poly-fluoroalkyl substances (PFAS) at historical aqueous film-forming foam (AFFF) sites presents a pressing issue. PFAS accumulation at air-water interfaces in the unsaturated media induces tension-driven flow (TDF), potentially altering interfacial tension (IFT). This phenomenon complicates the understanding, characterization, and modeling of PFAS transport in unsaturated conditions.
Previous research (e.g., Tschapek et al., 1981, 1984; Karkare et al., 1993; Karkare and Fort, 1996; Henry et al., 2003) has extensively investigated the effects of TDF, primarily focusing on the transport of single-component surface active agents as IFT sorbents. These studies have demonstrated that the presence of surfactants in unsaturated media leads to concentration-dependent surface tension gradients, causing capillary pressure gradients and moisture transition across the media (Karkare et al., 1993; Smith and Gillham, 1994 & 1999; Henry et al., 2002; Zeng & Guo, 2021; Abraham et al., 2022; Arora et al., 2022). While Costanza-Robinson and Henry (2017) integrated surfactant-induced flow into the HYDRUS-1D model, yielding substantial impacts on air-water interfacial areas, Brusseau et al. (2007, 2015, 2021) provided a contrasting perspective based on their unsaturated column experiments, suggesting a negligible influence on the same interfacial areas.
Given these diverse findings, our current study aims to contribute to the ongoing discourse by investigating and clarifying the role of surfactant-induced flow and its consequences on vadose-zone fluid behavior.
Approach/Activities
Our study employs bench-scale column experiments and numerical modeling to explore the implications of TDF on AFFF movement in the vadose zone. Tension-driven porewater redistribution was evaluated using the method detailed by Silva et al. (2019) for AFFF solutions, which was based on similar studies performed by Karkare et al. (1993) and Henry et al. (2001) for water-insoluble and water-soluble surface-active alcohols, respectively. The experimental setup consisted of preparing two samples (simulated groundwater and AFFF) of unsaturated sand for each half of the column.
This study also used the HYDRUS unsaturated flow and transport model (Šimůnek et al., 2016) that was modified to incorporate the effects of solution concentration-dependent air-water interfacial tension (IFT, also called surface tension) on solute transport in unsaturated porous media (Silva et al., 2019). The HYDRUS model was modified to capture the TDF effects on the transport of AFFF.
Results/Lessons Learned
AFFF demonstrated significant TDF, leading to notable lateral water movement and consequent shifts in moisture levels. The concentration-dependent TDF in unsaturated media induces substantial alterations in capillary pressure and hydraulic conductivity, up to 60% and two orders of magnitude respectively, causing significant flow disruptions compared to neglecting these IFT-reducing mechanisms. The adapted HYDRUS model effectively simulated the experiments with minimal calibration, providing insights into the dynamics of an AFFF source zone under TDF influence. Numerical analysis revealed that TDF could greatly expand the surface footprint of an AFFF source zone. Wetter initial conditions resulted in more spreading due to a larger AFFF reservoir, coupled with faster spreading owing to increased unsaturated hydraulic conductivity. This expansion occurs over decades, significantly impacting our understanding of relatively recent AFFF spills. Furthermore, the findings suggest that AFFF footprints may evolve over time due to changing climates, particularly prolonged droughts or wet periods, potentially yielding different source-zone footprints in various locales. These results enhance our conceptual grasp of AFFF source zones, crucial for precise site characterization, transport modeling, and mitigation strategies. Ultimately, this study reaffirms the proficiency of the modified HYDRUS model in replicating AFFF-induced TDF dynamics.