arrow
Return

Compact Solvation Enables Sub-Minute Sodium-Ion Storage: A Data-Driven Perspective

delete2026-06-10
delete0
PRE
AI
M
Mingxu Wang
C
Chenyu Tang
J
Jinyu Yang
Z
Ziyue Li
H
Hao Du
Q
Qin Li
H
Haoran Ji
X
Xinjie Li
Y
Yan Lu
方方 cover
方方 (Fang Fang)
M
Mao Su *
J
Jiafeng Ruan *
F
Fei Wang *
DOI:10.1002/adma.73658delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Extremely fast charging (XFC) of batteries holds significant importance in the era of intelligent technologies, yet the intricate role of electrolyte properties in determining XFC behavior remains obscure. Furthermore, conventional theories and in situ characterization techniques fails to elucidate electrochemical behaviors under sub-minute-level charging conditions. Herein, we report a data-driven approach for analyzing the effect of each independent physical and solvation property on sub-minute-level sodium-ion storage behavior. Causal graph analysis reveals that the size of the solvation clusters shows the strongest negative correlation with ultrafast Na+ storage in graphite. The optimized compact solvation electrolyte demonstrates an astonishing extreme current density of 250 A g−1 (corresponding to a power density of 46.78 kW kg−1graphite) and an unprecedented cycle life of 100 000 cycles. Most notably, the graphite||Na4Fe3(PO4)2P2O7 full batteries achieve stable sub-minute-level charging and discharging, exhibiting ultrahigh rate capabilities (up to 200 C, ∼3.4 s per charge) and ultra-stable cycle performance (24 000 cycles at 50 C, ∼29 s per charge). This work provides a promising pathway for the development of XFC battery, pioneering an innovative assessment strategy for next-generation electrolytes.
Keywords:
compact solvation
data-driven
extremely fast charge
graphite
Na+-solvent co-intercalation
sub-minute

Journal

Advanced Materials cover
Advanced Materials
IF:
26.8
Papers:
3.4W
Citations:
46.0W

Organization

F
fudan university
Scholars:
11.6W
Papers: 7.7W
Citations: 121
S
Shanghai Artificial Intelligence Laboratory
Scholars:
470
Papers: 258
Citations: 765