InterPore2026

Foam-Based Desorption of Multicomponent PFAS from Soil: Influence of Foam Generation Conditions

20 May 2026, 15:35

1h 30m

Poster Presentation (MS02) Environmental Porous Media: Water, Agriculture, and Remediation Poster

Speaker

Maxime Cochennec (BRGM)

Description

Per- and polyfluoroalkyl substances (PFAS) are ubiquitous environmental contaminants whose remediation in soils is challenging due to their amphiphilic nature, variable solubility, and resistance to degradation [1]. Although in-situ soil flushing has been investigated for PFAS-contaminated soils, conventional water-based approaches often require large water volumes and exhibit limited efficiency [2]. Because PFAS preferentially partition at air–water interfaces, foam-based flushing offers a promising alternative for mobilising PFAS sorbed to soil surfaces while reducing water consumption [3]. This study investigates the effects of foam pre-generator grain size and foam injection rate on PFAS desorption from a multicomponent contaminated soil.
A sandy soil composed of 92% sand, 5% clay, and 3% organic matter was used in this study. PFAS contamination was introduced using a multicomponent mixture prepared from an Aqueous Film Forming Foam (AFFF) stock solution, dominated by 6:2 fluorotelomer sulfonic acid (6:2 FTSA), 6:2 fluorotelomer sulfonamide betaine (6:2 FTAB), and 6:2 fluorotelomer sulfonamide (6:2 FTSaAM). Sorption experiments were conducted in water-saturated soil columns (30 cm length, 4 cm diameter) at a flow rate of 2 mL min⁻¹ for 17 pore volumes (PV). Foam was generated using sodium dodecyl sulphate (SDS) at 5× the critical micelle concentration and a foam quality of 90%. A 10 cm long foam pre-generator packed with fine sand was installed upstream of the contaminated soil column. Two pre-generator media with permeabilities of 35 and 105 Darcy were tested to achieve stable foam by controlling the pressure gradient. Once steady conditions were reached, foam was injected into the PFAS-sorbed soil column at flow rates of 3 and 9 mL min⁻¹ for up to 20 PV. Effluent samples were collected during sorption and desorption for chemical analysis.
Apparent viscosity measurements obtained during foam injection through the pre-generators revealed a shear-thinning flow behaviour. PFAS sorption results showed strong adsorption (>80%) for 8:2 FTSA, 10:2 FTSA, 6:2 FTAB, and 6:2 FTSaAM, with effluent concentration ratios (C₀/Cᵢ) ranging from 0.2 to 0.4 after 17 pore volumes (PV) of PFAS injection. In contrast, short-chain PFAS (PFBA and PFHxA) exhibited limited sorption, reaching C₀/Cᵢ values close to unity within 1.5 PV. Foam desorption experiments using the fine-grained pre-generator (35 Darcy) at a flow rate of 9 mL min⁻¹ resulted in recovery efficiencies of 64% and 74% for 6:2 FTAB and 6:2 FTSaAM, respectively, while recoveries for PFBA and PFHxA remained low (~0.1). Breakthrough curves showed peak C₀/Cᵢ ratios of 18 and 16 for 6:2 FTAB and 6:2 FTSaAM within the first 2.5 PV, followed by a decline to ~0.15 after 20 PV, indicating that most PFAS mobilisation occurred during the initial foam slugs. At a lower foam injection rate (3 mL min⁻¹), mobilisation efficiencies decreased to 0.50–0.57, with peak C₀/Cᵢ ratios of 17 and 10 reached within the first 4.7 PV. The reduced PFAS recovery at lower injection rates is attributed to changes in foam hydrodynamics under low-flow conditions, which reduce pressure gradients and limit the generation and renewal of air–water interfaces, thereby decreasing PFAS mobilisation.