Project Se(L)ect(i)vity (PTDC/EQU-EQU/6193/2020)

The recent growing demand for lithium worldwide, led by the Li-ion battery market, has sparked research into alternative sources of this material. In this context, selective lithium recovery from concentrated brines represents a sustainable and economical alternative to lithium mining activities. In this work, we developed a mathematical model of the recently implemented Lithium Membrane Flow Electrode Capacitive De-Ionization (Li-MFCDI) process, used to selectively extract lithium from a synthetic geothermal brine. The model was validated against the available experimental data and was used to perform a comprehensive parametric analysis. The model predicts the effects of the applied voltage, flow rates, and the adopted membranes on the process performance. These findings highlight the importance of the membrane conductivity-selectivity trade-off for process productivity. Furthermore, this simulation tool will substantially contribute to the development of this novel technology.

Power generation by reverse electrodialysis (RED) depends on ionic mass transfer through alternately arranged cation- and anion-exchange membranes. Chronopotentiometric measurements were carried out in an EDR-Z Mini stack (MEGA a.s.), equipped with Ralex heterogeneous membranes, separated by either sheet flow spacers or mesh-free gaskets. Various concentrations of model NaCl solutions were used to study the ohmic and non-ohmic resistances in the stack under different hydrodynamic conditions. In order to eliminate the uncertainties associated with a non-uniform distribution of the applied current over the membrane surface, the relaxation zone of the obtained chronopotentiograms was used to estimate the diffusion boundary layer thickness. It was found that this approach provide more accurate data under RED operating conditions, which are strongly influenced by entrance effects on mass transfer, especially for a spacer-free channel configuration. For shorter flow channels, since the salt concentration profile across the diffusion boundary layer is not yet fully developed, more power can be obtained than in the case of longer channels. The presence of spacers was found to reduce the boundary layer thickness, but also increased the ohmic resistance, due to their shadow effect over the membrane surface. The description of the impact of flow entrance effects on mass transfer, and therefore on diffusion boundary layer thickness, can be useful for characterization and further design and/or optimization of RED stacks performance.