Acquiring reliable hydrogen crossover data of hydrated ion exchange membranes to elucidate the ion conducting channel morphology

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Switzerland : 1996), 2023-09, Vol.471, p.144696, Article 144696
• Main Authors: Kim, SeungHwan, Song, Jaeheon, Duy Nguyen, Bao Tran, Lee, JongMyeong, Seong, JongGeun, Nam, SangYong, So, Soonyong, Kim, Jeong F.
• Format: Article
• Language: English
• Subjects: Green Hydrogen; Hydrogen Crossover; Ion Channel Morphology; Ion Exchange Membrane; Spherical Cluster Model; Water Electrolysis
• ISSN: 1385-8947; 1873-3212
• DOI: 10.1016/j.cej.2023.144696

•The hydrogen crossover data can be correlated to the ion channel morphology.•The hydrogen crossover data corroborates with the Gierke’s spherical cluster model.•It was shown that the ion channels exist as discrete spherical clusters.•A two-phase Maxwell model was derived to predict the hydrogen crossover accurately.•The ion channel connectivity can be deduced using the proposed method. The morphology of ion channels within ion exchange membranes (IEM) has long been disputed by researchers. Several accepted models include the Gierke’s discrete spherical cluster model, continuous parallel cylinder model, and flat-layered model. The primary experimental data to support each model were mostly small-angle X-ray spectroscopy (SAXS). In this work, we propose that hydrogen permeability (i.e. crossover) data of hydrated IEMs could provide valuable insights into understanding the ion channel connectivity and continuity within the membrane. We re-derived the two-phase Maxwell-Eucken transport models to predict the hydrogen crossover of hydrated IEMs. The model accurately predicted the hydrogen crossover of Nafion and in-house BPSH membrane series within 10% error range. The predicted results were experimentally validated using the pressure decay method (PDM) developed in our group specifically for this purpose. By comparing the degree of change in hydrogen crossover between the dry and hydrated membranes, it was possible to extract semi-qualitative information about the membrane’s controversial ion-conducting channel morphology. Interestingly, our experimental data and model prediction corroborates with the Gierke’s conjecture that the water phase within the ion channel exists as discrete spherical clusters, not as the continuous percolated phase (the parallel cylinder model). Although there is yet no consensus on the internal morphology of ion channels, our work suggests that hydrogen crossover data can be useful in elucidating the ion channel morphology.