Estimating PFAS Immobilization by Activated Biochars: Insights from Rapid Small-Scale Column Tests and Uncertainty Modelling

The increasing recognition of PFAS contamination in soil and groundwater necessitates effective and sustainable remediation strategies. Steam-activated biochars (SA-BCs), derived from renewable feedstocks (e.g., agricultural waste), are gaining attention as alternatives to conventional activated carbons for PFAS sorption due to their lower environmental footprint. However, studies on the sorption of PFAS to SA-BCs under field-realistic flow-through conditions remain scarce. Rapid small-scale column tests (RSSCTs) provide a versatile tool to estimate the performance of sorbents in full-scale systems (e.g., fixed-bed filter) in the lab. Yet, the influence of experimental uncertainties (e.g., varying flow rates) on the resulting breakthrough curves is often not discussed. Therefore, the aim of this study was twofold: (i) to investigate the performance of a SA-BC for the immobilization of PFAS in soil under flow-through conditions and (ii) to estimate the effect of experimental uncertainties on the fitting results.
Sorption of seven C4 to C8 per- and polyfluoroalkyl acids in synthetic groundwater to loamy sand amended with 0.5% SA-BC was investigated in small-scale columns (4cm x 1cm). Breakthrough curves were analysed using the ‘Ogata-Banks’ solution for the 1D ADE, including a Monte Carlo framework to assess the influence of experimental/ analytical uncertainties for flow rate, porosity and measured PFAS concentrations on the results.
Compared to untreated soil blanks, in which immediate breakthrough was observed for most compounds, the addition of 0.5% SA-BC led to a significant retardation with retardation factors of ~10 for PFBA to ~1500 for PFOS – with the elution order correlating with chain-length and functional group. Monte Carlo simulations for PFOS with relative uncertainties of 5% for flow rate, porosity and PFOS concentrations resulted in retardation factors of 854 to 3420 with a mean ± one standard deviation of 1545 ± 333. Corresponding breakthrough times for 50% PFOS range from 678 to 861 min, indicating that estimation of retardation and consequently breakthrough times based on single experiments could lead to substantial underestimation.
While our data highlight the potential of SA-BC for in-situ immobilization of PFAS in contaminated soils, they also emphasize the importance of considering experimental uncertainties. Ongoing work is focused on the influence of environmental factors, such as matrix composition, on reactive transport. Together, these advances will support a robust, uncertainty-integrated framework for getting deeper insights into the interaction of PFAS with SA-BCs and estimating PFAS immobilization in the field und varying conditions.