Modeling emerging contaminants breakthrough in packed bed adsorption columns by UV absorbance and fluorescing components of dissolved organic matter

This study investigated, using rapid small-scale column testing, the breakthrough of dissolved organic matter (DOM) and eleven emerging organic contaminants (EOCs) during granular activated carbon (GAC) filtration of different water qualities, including wastewater, surface water and synthetic water (riverine organic matter dissolved in deionized water). Fluorescing organic matter was better adsorbed than UV absorbance at 254 nm (UV254) and dissolved organic carbon (DOC) in all tested water. Furthermore, highest adsorption of DOM (in terms of DOC, UV254 and fluorescence) was observed during wastewater filtration. UV absorbing DOM had fast and similar breakthrough in surface water and synthetic water, whereas fluorescence breakthrough was very rapid only in synthetic water. PARAFAC modeling showed that different fluorescing components were differently adsorbed during GAC process. Particularly, fluorescing components with maxima intensity at higher excitation wavelengths, which are corresponding to humic-like fluorescence substances, were better removed than other components in all waters. As opposed to DOM, EOCs were better adsorbed during synthetic water filtration, whereas the fastest EOCs breakthrough was observed during filtration of wastewater, which was the water that determined the highest carbon fouling. Exception was represented by long-chained perfluoroalkylated substances (i.e., PFOA, PFDA and PFOS). Indeed, adsorption of these compounds resulted independent of water quality. In this study was also investigated the applicability of UV254 and fluorescing PARAFAC components to act as surrogates in predicting EOCs removal by GAC in different water matrices. Empirical linear correlation for the investigated EOCs were determined with UV254 and fluorescing components in all water qualities. However, fluorescence measurements resulted more sensitive than UV254 to predict EOC breakthrough during GAC adsorption. When the data from all water qualities was combined, good correlations between the microbial humic-like PARAFAC component and EOC removals were still observed and they resulted independent of water quality if considering only real water matrices (wastewater and surface water). On the contrary, correlations between EOC removals and UV254 removals were independent of water quality when combining data of surface waters and synthetic water, but a different correlation model was needed to predict EOCs breakthrough in wastewater.Accepted versio