Computational Discovery of Hydrogen Bond Design Rules for Electrochemical Ion Separation

Selective ion separation is a major challenge with far-ranging impact from water desalination to product separation in catalysis. Recently introduced ferrocene (Fc)/ferrocenium (Fc(+)) polymer electrode materials have been demonstrated experimentally and theoretically to selectively bind carboxylates over perchlorate through weak C-H center dot center dot center dot O hydrogen bond (HB) interactions that favor carboxylates, despite the comparable size and charge of the two species. However, practical application of this technology in aqueous environments requires further selectivity enhancement. Using a first principles discovery approach, we investigate the effect of Fc/Fc(+) functional groups (FGs) on the selectivity and reversibility of formate Fc(+) adsorption with respect to perchlorate in aqueous solution. Our wide design space of 44 FGs enables identification of FGs with higher selectivity and rationalization of trends through electronic energy decomposition analysis or geometric hydrogen bonding analysis. Overall, we observe weaker, longer HBs for perchlorate as compared to formate with Fc(+). We further identify F+ functionalizations that simultaneously increase selectivity for formate in aqueous environments but permit rapid release from neutral Fc. We introduce the materiaphore, a 3D abstraction of these design rules, to help guide next generation material optimization for selective ion sorption. This approach is expected to have broad relevance in computational discovery for molecular recognition, sensing, separations, and catalysis.