Symmetric Intercalative Desalination of Brackish Water Without Ion-Exchange Membranes

Ion-exchange membranes (IEMs) that separate cation-intercalation electrodes in symmetric Faradaic deionization (SFDI) increase the capital cost of desalination. We experimentally test SFDI cells that use nanofiltration (NF) membranes instead. Theory is first used to contrast the rate-dependent salt depletion in IEM-free SFDI with IEM-based SFDI. Theory reveals that salt removal scales directly with the intercalation-induced cation-depletion rate relative to the membrane's cation diffusion flux by including the following effects: a current efficiency λ for parasitic electrode processes and a permeance P_m,+ and transference number t_m,+ for membrane cation transport. Fitting to experiments indicates that bulk-electrolyte ion transport in the NF membrane's support layer (250 μm) dominates over active-layer ion transport (∼100 nm). An IEM-free SFDI flow cell using embedded, microinterdigitated flow fields was shown to desalinate feeds using 85 mM NaCl or 3.2 g/L of Instant Ocean synthetic salt. The cell produced either freshwater or drinkable water using 1.4-2.2 kWh/m³ at parity with reverse osmosis, electrodialysis, and membrane capacitive deionization. Water recovery was shown in IEM-free SFDI to increase with the charge transferred during batch-type experiments, contrasting its usual decrease with IEMs. Theory further suggests that NF membranes minimize concentration polarization relative to IEMs, decreasing mineral-scaling potential.