Accelerating anhydrous proton conduction via anion rotation and hydrogen bond recombination: a machine-learning molecular dynamics

Abstract

Phosphonic-acid-based electrolytes are key materials for anhydrous proton transport in fuel cells that are operatable at medium temperatures. However, these materials suffer from a severe tradeoff between proton conductivity and stability. Immobilizing phosphonic anion groups prevents anion leaching while suppressing proton transport. To reveal the origin of this relationship, we performed nanosecond-scale molecular dynamics simulations of phosphoric and phosphonic acids with different alkyl chains using machine-learned force fields. Simulations indicate that proton diffusivity is strongly correlated to the reorientation speed of anions. Thus, as the alkyl chain length increases, both the proton diffusivity and reorientation frequency decrease. Detailed analyses show that in all the materials, protons are shuttled between a pair of anions with a high frequency of approximately 10 ps⁻¹. However, only 0.1% of the shuttling protons are transported to the adjacent anion because of three orders of magnitude slower reorientation of anions that require recombination of H-bond network. Retaining the rotational freedom of anions is essential for enhancing anhydrous proton conductivity.