Characterization of humic acid reactivity modifications due to adsorption onto a-Al 2O 3

Abstract

Adsorption of purified Aldrich humic acid (PAHA) onto a-Al 2O 3 is studied by batch experiments at different pH, ionic strength and coverage ratios R (mg of PAHA by m 2 of mineral surface). After equilibration, samples are centrifuged and the concentration of PAHA in the supernatants is measured. The amount of adsorbed PAHA per m 2 of mineral surface is decreasing with increasing pH. At constant pH value, the amount of adsorbed PAHA increases with initial PAHA concentration until a pH-dependent constant value is reached. UV/Visible specific parameters such as specific absorbance SUVA 254, ratio of absorbance values E 2/E 3 and width of the electron-transfer absorbance band ? ET are calculated for supernatant PAHA fractions of adsorption experiments at pH 6.8, to have an insight on the evolution of PAHA characteristics with varying coverage ratio. No modification is observed compared to original compound for R=20mgPAHA/ga-Al2O3. Below this ratio, aromaticity decreases with initial PAHA concentration. Size-exclusion chromatography - organic carbon detection measurements on these supernatants also show a preferential adsorption of more aromatic and higher-sized fractions. Spectrophotometric titrations were done to estimate changes of reactivity of supernatants from adsorption experiments made at pH ˜6.8 and different PAHA concentrations. Evolutions of UV/Visible spectra with varying pH were treated to obtain titration curves that are interpreted within the NICA-Donnan framework. Protonation parameters of non-sorbed PAHA fractions are compared to those obtained for the PAHA before contact with the oxide. The amount of low proton-affinity type of sites and the value of their median affinity constant decrease after adsorption. From PAHA concentration in the supernatant and mass balance calculations, "titration curves" are experimentally proposed for the adsorbed fractions for the first time. These changes in reactivity to our opinion could explain the difficulty to model the behavior of ternary systems composed of pollutants/HS/mineral since additivity is not respected.