Doping proton transport channels in poly-electrolyte membranes with high acidic site density polymers

Abstract

The phase separated pore morphology of H1−xDx blends composed of host (H) and dopant (D) polymers in the presence of water is modeled by means of dissipative particle dynamics. H polymers contain a backbone of 35 connected hydrophobic A beads to which 5 branched A5[AC][AC]) side chains are non-uniformly distributed by attaching them to the last 5 backbone beads (C beads are acidic and hydrophilic). The C bead fraction (|C|) and ion exchange capacity (IEC) of the H polymer (|C| = 0.125; IEC=1.73mmol/cm3) are nearly 3 times less than of the D polymer (|C| = 0.33; IEC=4.6mmol/cm3). At low dopant volume fractions (ie. small x) the D polymers are distributed near the water containing pore interface and the Bragg peak position (near∼11nm) is hardly affected. With increase of x the radius of gyration of the H polymer backbone decreases and a second Bragg peak emergences near 2.5nm. This is explained by the large difference between <Nbond>host (=11.76) and <Nbond>dopant (=1.5), which are the average number of bonds (topological distance) that A beads are separated from a nearest C bead within the host and dopant architecture, respectively. MC tracer diffusion calculations through mapped morphologies predict a decrease of water diffusion with increase of D polymer volume fraction. Since the ion exchange capacity (proportional to|C|) increases with x, optimal blending ratios may be expected that favor proton conduction. These findings can be of special importance in the optimization of poly electrolyte membranes for fuel cell applications.