Minicolumn Test of Remaining GAC Performance for Taste and Odor Removal: Theoretical Analysis

Abstract

In drinking water treatment, it is difficult to predict the remaining service life of a partially spent granular activated carbon bed when contaminants are intermittent, such as taste and odor-causing compounds. A laboratory-scale minicolumn test using a grab granular activated carbon (GAC) sample from a full-scale bed was assessed using pore and surface diffusion model (PSDM) simulations. The impact of bed depth and flow velocity on the performance of preloaded GAC was negligible provided that the minicolumn test matched the same empty bed contact time as at full scale. The impact of minicolumn diameter was insignificant when the minicolumn to GAC particle diameter ratio was larger than 13. Minicolumns using a single representative grain size fraction were predicted to simulate full-scale beds with reasonable accuracy. The impact of temperature may be significant. A potential limitation of the test is that the media may be harvested at different GAC bed depths: accuracy requires that the adsorption capacity and kinetics of the adsorbate be relatively constant across those depths.