The development of high-performance fluoride-ion (F⁻) conductors is essential for the practical application of all-solid-state F⁻ batteries. To date, the crystal structures of the F⁻ conductors have been mainly limited to perovskite-, fluorite- and tyssonite-type fluorides. However, the room temperature conductivity of these F⁻ conductors is several orders of magnitude lower than that of cationic conductors. Furthermore, strategies to improve F⁻ conductivity are mainly limited to introducing defects in the crystal structure, which lags considerably behind the abundant means of conductivity enhancement in cationic conductors, such as inductive effects¹ and paddlewheel effects².

In mixed-anion compounds that containing more than one anion in the structure, the presence of anions with different polarization rates and electronegativity from F⁻ allows the construction of new crystal structures that cannot be obtained in single-anion compounds³. In this study, F⁻ conduction in fluorosulfide La₂SrF₄S₂ was investigated^[4]. In La₂SrF₄S_2, the triple-fluorite layer La₂SrF₄ allows for a two-dimensional F⁻ conduction, which is not found in single anion fluorides. The presence of a sulphur layer widens the conduction space of the F⁻ layer and weakens the interaction between cations and F⁻. Furthermore, we attempted to improve the F⁻ conductivity by cation-anion engineering. Specifically, (i) La₂SrF_4+xS_2-xCl_x, in which S²⁻ was changed to Cl⁻ to introduce excess F⁻, and (ii) La₂Sr_1-xPb_xF₄S₂, in which Sr²⁺ was changed to Pb²⁺ with 6s² lone pair electrons, were synthesized and the change in F⁻ conductivity were explored and explained from a crystallographic point of view.

(i) The ionic conductivity of La₂SrF_4.1S_1.9Cl_0.1 at 160 °C was 6.26 × 10⁻⁶ S cm⁻¹, which is five times higher than that of La₂SrF₄S₂. On the other hand, the activation energy of La₂SrF_4.1S_1.9Cl_0.1 is 0.51 eV, which is comparable to 0.53 eV of La₂SrF₄S₂. The increase in the pre-exponential factor can be attributed to the introduction of excess F⁻ in the lattice by replacing divalent S²⁻ with monovalent Cl⁻. This suggests that the use of mixed-anion compounds allows carrier engineering by anions, which is different from the carrier control of by cation substitution in previous F⁻ conductors.

(ii) The ionic conductivity of La₂Sr_0.6Pb_0.4F₄S₂ at 160 °C was 2.56 × 10⁻⁴ S cm⁻¹, which are both higher than those of La₂SrF₄S₂ and La₂SrF_4.1S_1.9Cl_0.1. The amount of F⁻ carriers and the F⁻-F⁻ distance did not change much before and after the introduction of Pb²⁺, suggesting that there are factors other than those previously considered responsible for the increased pre-exponential factor. ¹⁹F MAS NMR showed that the number of F⁻ in the hetero-cationic environment increased after the introduction of Pb²⁺, suggesting that the distorted cationic sublattice is partly responsible for the conductivity enhancement. Also, 6s² lone electron pairs in Pb²⁺ may also contribute to the increase in F⁻ conductivity.

In conclusion, using the mixed-anion compound La₂SrF₄S₂ as a model, we proposed new strategies to improve F⁻ conductivity by cation-anion engineering.