Simulations predict salt partitioning in nanostructured ion exchange membranes

Nanostructured ion exchange membranes exhibit an intricate network of aqueous pores that facilitate ion and water transport, which maybe useful in water purification and energy storage applications. In this study, we employ molecular dynamics simulations to investigate randomly sulfonated polystyrene-polymethylbutylene (PS-PMB) block copolymer membranes. We find that ion distribution within the pores is far from uniform and ions are prone to clustering. Both counterions and co-ions concentrate near the negatively charged pore wall, and their diffusive motion is correlated. To quantify the salt partition coefficient for these membranes, we utilize two approaches: (1) conventional “morphing” simulations, which calculate the free energy to transfer ions from bulk solution into the membrane, and (2) a novel technique to predict the equilibrium solution salinity from ion concentration profiles in the membrane. The two methods are consistent; both predict a partition coefficient exceeding that of the Donnan model, which assumes a uniform ion distribution within the membrane. We find that well-hydrated membranes with larger pores tend to preferentially absorb salt; to remedy this, cross-linking would be necessary to limit water uptake.