Predicting room-temperature conductivity of Na-ion super ionic conductors with the minimal number of easily-accessible descriptors

Given the vast compositional possibilities NaₙMₘMm'Si₃₋ₚ₋ₐPₚAsₐO₁₂, Na-ion superionic conductors (NASICON) are attractive but complicate for designing materials with enhanced room-temperature Na-ion conductivity σNa,300K. We propose an explicit regression model for σNa,300K with easily-accessible descriptors, by exploiting density functional theory molecular dynamics (DFT-MD). Initially, we demonstrate that two primary descriptors, the bottleneck width along Na-ion diffusion paths d₁ and the average Na-Na distance dNa-Na, modulate room-temperature Na-ion self-diffusion coefficient DNa,300K. Then, we introduce two secondary easily-accessible descriptors: Na-ion content n, which influences d₁, dNa-Na, and Na-ion density ρNa; and the average ionic radius rM of metal ions, which impacts d₁ and dNa-Na. These secondary descriptors enable the development of a regression model for σNa,300K with n and rM only. Subsequently, this model identifies a promising yet unexplored stable composition, Na2.75Zr1.75Nb0.25Si₂PO₁₂, which, upon DFT-MD calculations, indeed exhibits σNa,300K > 10³ S·cm⁻¹. Furthermore, the adjusted version effectively fits $140$ experimental values with R²=0.718.