Computer simulation of activated carbon adsorption for multi-component systems

Abstract

Granular activated carbon (GAC) adsorption has been commonly used for removing synthetic organic chemicals (SOCs) from groundwater. The removal efficiency is largely dependent on the interactions of competing organic chemicals which are present in the solution. In order to design a GAC adsorption system, the information regarding the removal capacity of a solute in the presence of other competing organics is required. Such information can be gained empirically from pilot studies, but tests of this nature are typically lengthy and time-consuming. To tackle this problem and also to provide useful adjuncts to pilot-scale investigations, a simulation model for carbon adsorption is hereby developed based on the ideal adsorbed solution theory (IAST) and homogeneous surface diffusion model (HSDM). The incorporation of IAST into the kinetic HSDM model makes it possible to predict the effluent concentration of SOCs in a multi-component adsorption system. To verify the applicability of IAST on the activated carbon adsorption, laboratory study of batch adsorption tests were conducted to investigate the removal of phenolic compounds from aqueous solution for both single and multi-component systems. Based on the comparison of the experimental results with the predicted data from IAST, it indicates that IAST gives a general satisfactory prediction of multi-component competitive adsorption equilbria using their respective single component isotherm for each chemical. Moreover, the simulation model (IAST + HSDM) has been found to have a good fit to the Noll et al. (1991) experimental data for two-component adsorption column tests under a various range of initial concentrations and flow rates. Sensitivity analysis of a two-component system consisting of phenol and p-chlorophenol shows a significant deviation in predicted breakthrough curve, with the Freundlich K value being deviated by ±20%. This reflects the importance of the Freundlich K in achieving a reliable modelling simulation.