Role of cationic groups on structural and dynamical correlations in hydrated quaternary ammonium-functionalized poly(p-phenylene oxide)-based anion exchange membranes

Extensive atomistic molecular dynamics (MD) simulations employing a polarizable force field have been conducted to study hydrated anion exchange membranes comprised of a poly(p-phenylene oxide) (PPO) homopolymer functionalized with quaternary ammonium cationic side groups and hydroxide anions. Representative membranes with different cationic structures have been investigated to study correlations between polymer architecture, morphology and transport properties of hydrated membranes. Specifically, hydrated polymers with five different quaternary ammonium cationic groups (R1: –CH3, R2: –C2H5, R3: –C3H7, R4: –C6H13 and R5: –C4H8OCH3) and degree of functionalization of 50% were investigated at three hydration levels (λ = Nwater/Ncation = 5, 10 and 17). Effects of the polymer structure on the distribution of water-rich domains and dynamic relaxations were systematically investigated to uncover the complex interplay between the degree of hydrophobicity/hydrophilicity of the cationic groups, morphology, connectivity of water domains, and the hydroxide transport mechanisms. Structural and dynamical analysis indicates that the bottlenecks, formed between the water-rich domains, create a substantial free energy barrier for hydroxide transport associated with the partial loss of anion hydration structure. The energy penalty associated with the loss of the hydration structure hinders the vehicular transport of the hydroxide anion. The optimal structure of functionalized homopolymer chains should be sufficiently hydrophobic to create nanophase segregation and form an interconnected network of water channels with a minimal amount of narrow bottlenecks that inhibit the vehicular motion of hydrated anions. We demonstrate that utilization of asymmetrically modified cationic groups is a promising route to achieve the desired water channel morphology at low hydration levels.