Résumé
NaSICON electrolytes, such as Na1+xZr2(SiO4)x(PO4)3–x (NZSP), constitute promising candidates for solid-state battery (SSB) development. Research on such fast superionic conductors has primarily focused on two key phenomena acting specifically on Na+ ion migration: (i) the Na-concentration-driven modulation effect and (ii) the incidence of substitution. While numerous experimental and computational studies have established the fundamental role of concerted migration in ionic conduction, the precise influence of bottleneck size along with its dependence on NaSICON composition remains elusive. In view of participating in this research field and following an experimentally tested strategy, suggesting that the migration bottleneck can be expanded by partially substituting diffusing Na+ ions with larger-radius alkali elements, we investigated the impact of the introduction of such point defects (i.e. K+ or Cs+ replacing Na+) on structural and Na+ diffusion aspects in the NZSP crystal structure. A proof of concept of the interest linked to this unconventional doping approach has been searched for. Theoretical investigations relying on density functional theory (DFT) and subsequent kinetic Monte Carlo simulations were involved to unravel interrelations between the ionic radius of the substituting ion and bottleneck sizes, structural changes, diffusion pathways, and ionic conductivity features. Apart from an opening of the bottleneck along the migration path as a common feature, a clear differentiation between both kinds of substituents was evidenced on various aspects, K-NZSP outperforming the undoped counterpart and effectively enabling the maximization of ionic conductivity in these envisaged NaSICON-type matrices. Furthermore, the identification─emerging from this study─of a critical bottleneck size in such systems may contribute to provide a further key clue and lead to well thought-out crystal chemical engineering of improved materials for this research area.