Monitoring internal electrolyte decomposition byproducts is pivotal for the early warning of thermal runaway in lithium-ion batteries yet remains a formidable challenge due to the harsh chemical environment. Herein, we engineer a robust electrochemical sensor based on a CsPbBr3/Al2O3@EVA heterojunction architecture to achieve real-time, in situ tracking of lithium methoxide (CH3OLi) evolution. Through interface engineering, the synthesized quantum dot−Al2O3 composite (8.7at%Al) achieves exceptional stability in reducing electrolytes, attributed to the synergistic dual-passivation of Br−Al interfacial bonding and EVA encapsulation. The sensor exhibits superior sensitivity (0.3 at 10% CH3OLi) and rapid kinetics (response/recovery: 8.98 s/56.11 s), driven by the promoted molecular diffusion in mesoporous Al2O3 and polarized adsorption within the Pb+2−Br− framework. Density functional theory (DFT) calculations further corroborate this mechanism, revealing a strong Li+ binding energy of −1.54 eV at Al−O−Br active sites. Furthermore, by integrating the sensor signals with an XGBoost machine learning algorithm (accuracy >99%), we demonstrate a smart monitoring system capable of accurately predicting battery voltage variations and identifying potential safety hazards. This work establishes a new paradigm merging interface-stabilized materials with intelligent algorithms, transforming battery safety management from passive protection to active early warning.
Mu et al. (Thu,) studied this question.