Open-Circuit Potential Drift in Intercalation Electrodes: Role of Charge Redistribution in a Prussian Blue Analog

Prussian blue analogs (PBAs) are used as electrode materials in energy storage and water deionization cells due to their reversible cation intercalation capability. Despite extensive research on their performance and intercalation mechanisms, little attention has been given to their behavior under open-circuit conditions. Recent studies using symmetrical PBA electrodes in two electrode deionization cells reported that after constant current cycling in dilute NaCl (<0.2 M), the cell voltage dropped under open-circuit conditions, which substantially increased the amount of energy consumed for deionization. However, it remains unclear which electrode (anode/cathode) experienced potential drift and if it was influenced by the low salinity of the electrolyte. Here, we performed a series of electrochemical experiments under different charging and discharging regimes and electrolyte compositions to determine the processes that contributed most significantly to open-circuit potential drift. The data indicated that charge redistribution within the electrode was the main contributor to open circuit potential drift, with electrode dissolution and parasitic reactions playing negligible roles. A one-dimensional finite element model was constructed to simulate charge redistribution by accounting for cation diffusion under open-circuit conditions. The open-circuit potential profiles generated by the model were validated against experimental trends, confirming the occurrence of charge redistribution. A Monte Carlo analysis of the model was conducted to determine the relationship of potential drift to key factors such as applied current, electrode thickness, diffusion coefficient of intercalating ions, and intercalation capacity. Subsequently, a dimensionless number (Da) was developed based on the Dahmköhler number to relate the extent of potential drift resulting from combinations of these factors. The analyses revealed a strong positive correlation between simulated potential drift and Da. Among the key factors studied here, the diffusion coefficient and applied current had the largest impact on Da and, consequently, on potential drift.