Why Chemical Stability of Anion-Exchange Membranes Depends on the Hydration of the Aemfcs

Abstract

Anion-exchange membrane (AEM) fuel cells (AEMFCs) can potentially revolutionize energy storage and delivery; however, their commercial development is hampered by the chemical decomposition of the anion exchange membranes during operation. As the current density in the AEMFC increases, more water is consumed at the cathode side, reducing the hydration level and, in turn, increasing the nucleophilicity of hydroxide anions and its capability to attack the quaternary ammonium (QA) groups.1–3 Several new QA salts and ionomers have been proposed to address this challenge, but while they perform well in ex-situ chemical studies, their performance is very limited in real AEMFC studies.In this study we experimentally show that water concentration in the environment of the hydroxide anion significantly impact its reactivity. In particular, we compare different QA salts that have been previously studied and test their alkaline stability at different hydration levels, as well as different temperatures. We measure the kinetics of the hydroxide-QA chemical reaction and use the data as an experimental base to understand how microsolvation affects the hydroxide reactivity – as a nucleophile, a base, and a reducing agent.4,5 The results demonstrated a disparity in the water microsolvation effect of hydroxide through E2 and SN2 kinetic experiments with two stable QA salts, trimethylbenzyl ammonium bromide (TMBA) and 6-azonia-spiro [5,5] undecane bromide (ASU) – at high hydration levels, when hydroxide first solvation shell is fully hydrated, ASU is more stable than TMBA. However, at low hydration levels, when hydroxide ion presents low microsolvation, TMBA is significantly more stable than ASU (figure 1).6 The reaction rate constants indicate that the ASU reaction kinetics, which is a consequence of hydroxide basicity, is significantly more affected by the change in microsolvation than TMBA reaction kinetics, which depends on hydroxide nucleophilicity. These results have important implications for long term operation of AEMFCs. Figure 1. TMBA and ASU concentrations as a function of time in 0.6 M OH- in DMSO-d6 solutions for hydration level λ = 1 and 8.