Fundamental kinetic constants for breakthrough of per- and polyfluoroalkyl substances at varying empty bed contact times: Theoretical analysis and pilot scale demonstration

Abstract

Successful treatment of per- and polyfluoroalkyl substances (PFAS) contamination by adsorptive fixed bed media has been well established in literature and practice, but kinetic effects are important for these compounds. Common kinetic models of adsorption are descriptively fit to experimental data obtained from a particular design case, leading previous literature to dismiss these models as unable to provide useful information across changes to key operational variables. However, some descriptive models, such as the Clark and Bohart-Adams models, have mechanistic derivations with fundamental constants. The objective of this study was to evaluate the consistency of these fundamental constants across different experimental empty-bed contact times (EBCTs), and to evaluate the accuracy of using fundamental constants obtained at one EBCT to predict breakthrough at another EBCT.A total of 53 pilot-scale PFAS breakthrough curves through granular activated carbon (GAC) and ion exchange (IX) media were fitted to the Bohart-Adams and Clark models. Models were used to generate fundamental bed capacity, Freundlich, and kinetic constants. For each solute-media pair, the fundamental bed capacity and Freundlich constants had good agreement and low error across different EBCTs, while the kinetic constant had higher error with some EBCT dependance, but good order-of-magnitude agreement. The fundamental constants provided insights into the solute-media interaction, and predictive curves based on these fundamental constants provided reasonable estimates of breakthrough. The measurement of consistent values of underlying constants measured at different EBCT in this study suggests that the Clark and Bohart-Adams models can indeed be used to make predictions by varying parameters that are explicitly accounted for in the models, allowing for breakthrough curves generated at one EBCT to be used to predict breakthrough at another EBCT.