Dynamic modeling of the biological activated carbon (BAC) process and its experimental corroboration

In biological activated carbon (BAC) systems, organic pollutants are removed from the liquid phase by adsorption on granular activated carbon (GAC) and/or by biodegradation in the biofilm coated on GAC surface. Although the mechanisms for organic removal in such systems and the relationship between carbon adsorption and biofilm growth have been studied extensively, implementation of full scale systems has been limited. To advance BAC system design capabilities, better quantification of BAC system performance under steady-state and transient operational conditions is needed. In this research, the BAC process was studied through mathematical model simulations and experimental investigations.; A mathematical model was developed for expanded bed BAC systems with single rate limiting carbon source for carbon adsorption and biodegradation. Carbon adsorption was described by an empirical model developed by Radke, which fits adsorption isotherms in wide concentration ranges, while intraparticle mass transfer in GAC particles was modeled by homogeneous solid diffusion. Biodegradation was described by Monod kinetics. To make the model more general and complete, biofilm was considered as a shell growing on spherical GAC particles; biofilm accumulation was considered as the balance of biological simulation and hydraulic shear loss; and hydraulic shear loss rate was expressed as a function of hydraulic shear stress acting on GAC/biofilm particles. In addition, the change of solid phase volume in the system as a result of biofilm growth also was considered in the model. To improve the efficiency and stability of numerical solution, a mathematical transformation was performed to the governing equations of GAC/biofilm particles and Galerkin's finite element method was used to solve the transformed general equation. With this new solution approach the GAC/biofilm model was solved without iterations in each time step as required by previously used numerical methods. The model was able to simulate organic substrate removal, biofilm growth and suspended biomass growth in BAC systems as a dynamic process.; Experimental investigation of BAC process were conducted in 1-inch and 2-inch column systems. Effluent substrate concentration, biofilm thickness and density, and suspended biomass growth were studied under initial startup, steady-state, and transient conditions. At loading rates of 0.007 to 0.026 g phenol/g GAC-d, complete removal of phenol was achieved in BAC columns with very low effluent DOC. The BAC systems were able to sustain 5 fold and 10 fold phenol shocks with relatively low phenol breakthrough. Biofilm density was found to decrease with the increasing biofilm thickness. This biofilm density change was believed a result of biofilm population shifts and nutrient limitations in thick biofilms.; With the experimental data collected from BAC columns and the parameters of biokinetics and carbon adsorption obtained from separate experiments, the BAC model developed was calibrated and corroborated.