Formatted Title
Smoldering Treatment of PFAS: Part II. Field Demonstration
Background/Objectives
A pilot field-scale demonstration of smoldering combustion was conducted at a Canadian military base located in Eastern Ontario, Canada, using soils from the soil containment cells in the former firefighting training area (FFTA). This study was conducted under the U.S. Department of Defense (DoD) Strategic Environmental Research Program (SERDP) and explored the application of smoldering combustion to treat per- and polyfluoroalkyl substances (PFAS)-impacted soils and media. The project objective was to demonstrate that the laboratory results in the abstract provided for this conference by Harrison et al. can be scaled to real-world conditions.
Approach/Activities
Smoldering is a self-sustaining (i.e., energy efficient), flameless form of combustion that can use low volatility, high energy contaminants as fuel to support their own destruction. PFAS do not support smoldering combustion in and of themselves, and their high thermal stability often requires temperatures greater than 900°C for destruction; however, a supplementary fuel can be used to reach these temperatures in a smoldering combustion reaction to facilitate the destruction of PFAS.
Laboratory work demonstrated that smoldering temperatures >900°C are achieved by adding small amounts of granular activated carbon (GAC), which may also contain PFAS. The addition of calcium oxide (CaO) improved the destruction of PFAS while minimizing the release of hydrogen fluoride (HF) and volatile fluorinated organic compounds, with a high degree of mineralization of PFAS and mass recovery of organic fluorine.
A field pilot was conducted to verify the laboratory results and assess the formation of PFAS destruction byproducts during two pilot-scale applications of smoldering. The study involved the smoldering treatment of two 10 m3 batches of PFAS-contaminated soils that were amended with GAC and CaO. In addition to collecting and analyzing the pre-and post-treatment soils, HF, PFAS trapped in emissions trains, and silonite canisters were also used to collect gaseous samples for targeted and non-targeted PFAS analysis. Unlike the laboratory study that used known amounts of spiked PFOS absorbed on GAC and then smoldered in a porous medium, this study used virgin GAC amended to a field soil with unknown PFAS mass that was incompletely characterized. This is the first field pilot test demonstrating smoldering to treat PFAS in contaminated soils.
Results/Lessons Learned
The field testing confirmed the laboratory results and demonstrated that the process can be scaled to treat PFAS-impacted soils. PFAS concentrations in the post-treatment soils were non-detect or below their U.S. EPA’s Regional Screening Levels for residential soils. Fluorine liberated from the destruction of PFAS was associated with calcium fluoride (CaF2) on soils instead of being converted to HF with minimal products of incomplete destruction (PID) or short-chain PFAS generation. This is consistent with only 1% of the estimated initial organic fluorine mass being found as trapped HF and approximately 0.2% of the organic fluorine mass associated with detected PFAS found on the GAC traps used to capture emissions. Suspect screening of gas collected in silonite canisters found six fluorinated compounds, four of which were also detected in laboratory blanks. The two unique fluorinated compounds were not PFAS but long-chain (>20 C) compounds that had one or two fluorines associated with phenyl groups.