UNDERSTANDING AND MODELING THE SORPTION ON ANION EXCHANGE RESINS USING POLY-PARAMETER LINEAR FREE-ENERGY RELATIONSHIPS AND PHASE CONVERSION

Abstract

Priority organic and emerging contaminants are a growing concern for drinking water treatment due to their increasing presence in the environment. This study developed a predictive model for the sorption of anionic organic contaminants from drinking water on three anion exchange resins: a strong polystyrenic (IRA-910), weak polystyrenic (IRA-96), and a strong polacrylic (A860). The model quantifies the individual mechanisms of sorption using poly-parameter linear free energy relationships (pp-LFERs) and the feasibility of phase conversion (e.g., an ideal gas phase as the reference state) for ionic species was examined. To develop the model, a training set of isotherms was obtained using aliphatic and aromatic carboxylates, phenols, anilines, nitrobenzene, and ibuprofen. These compounds were chosen as model organic contaminants in the environment. The training set and 1-3 test compounds (3-methyl-2-nitrobenzoate, phenol, and 4-nitroaniline) were accurately predicted using the created model for each resin. An understanding of the effects of resin structure on sorption interactions was also developed that focused on ionic functional groups, resin matrix, and hydrophilicity (i.e. water content). It was shown that greater sorption efficiency was achieved when electrostatic (ion exchange) and nonelectrostatic (adsorption) interactions were present together to create a synergistic addition. However, sorption on ion exchangers was poor if the pH of the system approached levels lower than the sorbate pKa. Additionally, weak base exchanges lose exchange capacity as pH levels approach resin pKa (IRA-96 pKa = 6.0). Additional contributions to the sorption mechanisms were observed by studying various electron donating/withdrawing functional groups on the contaminants. It was concluded that π-π and H-bonding interactions contributed a greater amount to the nonelectrostatic mechanisms than cavity formation forces and nonspecific forces. A comparison between the three resins showed that IRA-96 (weak base polystyrenic) had a greater removal capacity than IRA-910 (strong base polystyrenic), followed far behind by A860 (strong base polyacrylate). This is due to differences between the resins, such as the hydrophilicity, the density of the ion exchange group, and the presence of aromatic rings within the matrix structure. Although the modeling method accurately predicted the phase change from aqueous to sorbent phases, it was shown that the SPARC calculated aqueous-gas ion transfer energies were poor estimations of the transfer energy to the ideal gas phase and further study is necessary to accurately determine this value. This modeling methodology is believed to be applicable to emerging contaminants such as pharmaceuticals in water systems and helps further new water treatment technologies while developing a mechanistic understanding of electrostatic and nonelectrostatic interactions in general. This can be applied to additional separation processes such as chemical purification and chromatographic separation.