Colloidal Activated Carbon for In Situ Remediation of Perfluorinated Alkyl Substances: Effects of Groundwater Chemistry and Implications for Technology Use and Longevity

Per- and polyfluoroalkyl substances (PFAS) are a family of persistent anthropogenic compounds that are ubiquitous in the environment. PFAS are found at many US Military sites and airports due to their presence in aqueous film-forming foams (AFFF) which are used for firefighting activities. PFAS impacted groundwater has traditionally been managed with ex situ pump-and-treat systems, but a rapidly emerging technology is in situ sequestration using colloidal activated carbon (CAC). In situ adsorptive barriers are created by direct injection of a CAC suspension into an aquifer where it deposits on soil particles. When placed within or downgradient of a PFAS plume, CAC barriers slow contaminant migration and decrease mass flux. However, similar to ex situ adsorption, in situ CAC will eventually reach an inherent adsorption capacity at which breakthrough may occur. Long-term barrier performance remains uncertain under changing water chemistry scenarios and for a range of PFAS structures. A systematic evaluation of how groundwater constituents impact CAC longevity is required to inform current remedial practice and guide future barrier design. The work presented in this thesis provides a comprehensive evaluation of the factors affecting the performance of CAC as an in situ remedial technology. Results presented provide practical information to remediation professionals and advance the fundamental understanding of the mechanisms governing sorptive PFAS interactions with activated carbon surfaces.

The first objective presented in Chapter 2 is to quantify the effects of common groundwater solutes on PFAS adsorption to CAC and assess their impact on remedy longevity. Batch adsorption tests were used to quantify the extent of PFAS adsorption to a commercial CAC in the presence of inorganic anions, dissolved organic matter (DOM) and stabilizing delivery polymer. Adsorption results were interpreted in light of PFAS structural characteristics and CAC surface characterization to propose mechanisms of adsorption. Inorganic anions decreased the adsorption of short-chain PFAS (<7 perfluorinated C) while the adsorption of long-chain PFAS (≥ 7 perfluorinated carbons) was moderately enhanced. DOM decreased the adsorption of all PFAS in a chain length dependent manner where high DOM concentrations (10 mg/L) decreased adsorption of perfluorooctanoic acid (PFOA, long-chain) by a factor of 2 and completely hindered adsorption of perfluorobutanoic acid (PFBA, short-chain). Low MW DOM caused more significant decreases in PFAS adsorption than high MW DOM as it is expected to block access to CAC micropores more significantly. Anionic delivery polymer, carboxymethyl cellulose (CMC), had negligible impact on PFAS adsorption. Longevity modeling demonstrated that groundwater solutes limit the capacity for PFAS in a CAC barrier, particularly for short-chain PFAS.

The second objective presented in Chapter 3 was to evaluate the performance of two CAC materials for PFAS adsorption in AFFF-impacted groundwater and determine which groundwater variables are most influential to adsorption performance. PFAS adsorption was quantified for two CAC materials with different surface chemical properties in four groundwaters collected from AFFF-impacted sites with varying chemistries. PFAS adsorption was found to be inhibited in each groundwater with short-chain adsorption more affected than long-chain. Groundwater with high concentration of total organic carbon (TOC), diesel range organics (DRO), and co-occurring PFAS had the highest competitive adsorption effects. Correlation analysis validated that TOC and DRO were the groundwater variables most strongly correlated with decreased CAC performance. Comparison of two CAC materials also showed that CAC with a high pHPZC (9.5) was more advantageous for PFAS adsorption compared to the other material (pHPZC = 6.7) due to increased electrostatic interaction capacity with anionic PFAS, but this advantage was minimized in groundwater with high ionic strength due to charge screening effects. Targeted modifications to an unimpacted water were also performed and adsorption results showed that TOC is the most impactful variable for long-chain PFAS, but an additive effect of groundwater variables is more likely for short-chain PFAS.

The third and final objective presented in Chapter 4 was to evaluate the potential for PFAS release from in situ barriers by evaluating the extent and timing of PFAS desorption from CAC due to changes in water chemistry. One-dimensional flow-through column studies were used to quantify the release of four representative PFAS from CAC due to changing influent chemistry conditions. PFAS release under control conditions (1 mM NaHCO3, pH = 7.5) varied from <1% for PFOA, 30% for perfluorobutane sulfonamide (FBSA), 46% for perfluorobutane sulfonic acid (PFBS), to 83% for perfluoropentanoic acid (PFPeA) and hydrophobic interaction capacity determined the extent of release. Elevated ionic strength decreased mass of PFOA released and only slightly increased the release of the three shorter-chain PFAS. High IS also decreased the rate of short-chain PFAS release. Low MW DOM led to higher displacement of all PFAS compared to high MW DOM and caused PFAS mass release at later eluting pore volumes. When influent contained DRO, the extent of PFOA release was significantly increased where 70% of previously sequestered mass was recovered in effluent. Short-chain release was not as significantly increased in the presence of DRO. Results were interpreted considering both CAC surface chemistry, PFAS structural characteristics, and mass transfer kinetics to propose mechanisms of displacement from the CAC surface.

Chapter 5 details the contributions of this work and outlines perspectives for future work in this research area. Results immediately benefit practitioners and will aid in technology design and longevity modeling efforts. The mechanistic interpretations provided will also support scientists and engineers studying PFAS adsorption and fate and transport behavior. Future research directions include designing barriers with multiple types of adsorbents, creation of new engineered adsorbent materials, development of a predictive tool to estimate PFAS adsorption based on groundwater and carbon characteristics, and finally exploring the potential for PFAS mass destruction in situ. Overall the work presented in this thesis advances both the practical and fundamental understanding of in situ colloidal carbon for the remediation of PFAS-impacted groundwater.