Screening Fluoride-Ion Conductors for Improved Batteries

Lithium-ion batteries are widely used, e.g., in portable electronics. Alternative battery technologies could help to achieve higher capacities and voltages. Fluoride-ion batteries (FIBs), for example, have the theoretical potential to outperform lithium-ion batteries. However, they have been hampered so far by the insufficient transport properties and stability of the few existing fluoride-ion conductors. The discovery of improved fluoride-ion conductors is, thus, important.

Scott C. Warren, The University of North Carolina at Chapel Hill, USA, and colleagues have used a new approach for materials discovery for the high-throughput computational screening of fluoride-containing materials to find promising fluoride-ion conductors. The team screened 9747 fluoride-containing materials from the Materials Project database and identified twelve systems with high fluoride mobility.

The researchers call their method a “decoupled, dynamic, and iterative (DDI) framework”. This indicates that the necessary calculations are decoupled and can be performed in any order, the computational resources are assigned dynamically, and the models are updated in an iterative manner. In a traditional, hierarchical approach, simplified models are used first to screen materials. Promising candidates are then investigated with more precise, but more costly computational methods. In the DDI framework, in contrast, the team produced a growing set of high-quality calculations using a database of candidates with a dynamic priority ranking. This allows them to assess the suitability of lower-cost calculations, refine the underlying models, and reallocate resources to determine fluoride conduction properties quickly and accurately.

The identified candidate materials include some well-known fluoride-ion conductors such as PbF₂ or LaF₃, but also compounds such as ZnTiF₆ and MgTiF₆, which had not been synthesized in their dehydrated form so far. The full list of promising materials could serve as a jumping-off point for further research.