Abstract
<jats:p>The desolvation kinetics of Na+ at the electrode-electrolyte interface critically limit the wide-temperature performance of sodium-ion batteries (SIBs). Conventional radial distribution function analyses, which rely on spherical symmetry averaging, inherently fail to capture the microscopic conformational inhomogeneity that governs this process. Herein, we develop an automated conformational clustering framework that integrates unsupervised machine learning with a multiscale DFT-MD approach to resolve the temperature-dependent solvation evolution in a model carbonate-based sodium-ion electrolyte. By mapping high-dimensional trajectory data, this approach quantitatively identifies solvent-separated ion pairs (SSIP) and contact ion pairs (CIP) conformations, revealing a temperature-driven population shift from SSIP-dominated (79.9% at 233.15 K) to CIP-dominated (61.7% at 333.15 K). Density functional theory vertical energy calculations demonstrate that elevated temperatures substantially lower the SSIP-to-CIP conversion barrier (from 98.3 kcal/mol to 47.0 kcal/mol). Consequently, the increased CIP proportion reduces the system-averaged desolvation binding energy from 72.6 kcal/mol to 67.6 kcal/mol, which directly accounts for the order-of-magnitude enhancement in sodium cation mobility observed in molecular dynamics simulations. This work establishes a quantitative mechanistic closed loop linking microscopic conformational transitions to macroscopic transport kinetics, providing explicit molecular design principles for advanced wide-temperature SIB electrolytes.</jats:p>