A rapid kinetic dye test to predict the adsorption of 2-methylisoborneol onto granular activated carbons and to identify the influence of pore volume distributions

Abstract

The authors have developed a kinetic dye test protocol that aims to predict the competitive adsorption of 2-methylisoborneol (MIB) to granular activated carbons (GACs). The kinetic dye test takes about two hours to perform, and produces a quantitative result, fitted to a model to yield an Intraparticle Diffusion Constant (IDC) during the earlier times of dye sorption. The dye xylenol orange was probed into six coconut-based GACs and five bituminous-based GACs that hosted varied pore distributions. Correlations between xylenol orange IDCs and breakthrough of MIB at 4 ppt in rapid small-scale column tests (RSSCTs) were found with R2s of 0.85 and 0.95 for coconut carbons that processed waters with total organic carbon (TOCs) of 1.9 and 2.2 ppm, respectively, and with an R2 of 0.94 for bituminous carbons that processed waters with a TOC of 2.5 ppm. The author sought to study the influence of the pore sizes, which provide the adsorption sites and the diffusion conduits that are necessary for the removal of those compounds. For coconut carbons, a linear correlation was established between the xylenol orange IDCs and the volume of pores in the range of 23.4–31.8 Å widths (R2 = 0.98). For bituminous carbons, best correlation was to pores ranging from 74 to 93 Å widths (R2 = 0.94). The differences in adsorption between coconut carbons and bituminous carbons have been attributed to the inherently dissimilar graphene layering resulting from the parent materials and the activation processes. When fluorescein dye was employed in the kinetic dye tests, the correlations to RSSCT-MIB performance were not as high as when xylenol orange was used. Intriguingly, it was the same pore size ranges that exhibited the strongest correlation for MIB RSSCT's, xylenol orange kinetics, and fluoroscein kinetics. When methylene blue dye was used, sorption occurred so rapidly as to be out of the scope of the IDC model.