It has proven effective to recover metal compounds from aqueous mixtures by use of antisolvents; organic compounds that induce selective precipitation. A challenge with antisolvents is that they are both costly to produce and recover on an industrial scale. In recycling of lithium-ion batteries and recovering critical metals, we find that electrodialysis can be a competitive method for purifying and recycling antisolvents. In this study we investigate the use of electrodialysis to separate salt and water from a ternary solution of water, KCl and ethanol. A coupled non-equilibrium electrochemical model is developed to understand how such systems may be operated, designed, and which characteristics that are required for the ion exchange membranes. We demonstrate how the water transference coefficients of the membranes should be tuned in the process optimisation and why membrane property design is crucial to the success of this concept. Residual mixtures from antisolvent precipitation, with ethanol (EtOH) solvent weight fractions around 0.6-0.7, can be demineralised and the EtOH fraction increased by 0.1-0.2 at an energy requirement of 60-200 kWh mEtOH−3 by use of electrodialysis. In an example application of the concept, aqueous KCl is precipitated by recycled ethanol in a cyclic process, requiring 0.161 kWh molKCl−1. This example case considers complete ethanol rejection by the membranes and abundant water co-transport, characterised by the transference coefficients: tw=15 and ta=0 for water and EtOH respectively. The findings pave the way for new applications with aqueous mixtures of critical metals.