Electrochemical energy storage plays a vital role in enabling the integration of renewable energy and supporting the development of smart grids. Among various technologies, lithium-ion batteries have emerged as the most promising solution due to their high energy density, fast charging capability, long cycle life, and overall reliability. However, despite these advantages, lithium-ion batteries face significant safety concerns, with thermal runaway being the most critical. The use of flammable organic solvents in conventional electrolytes is a major contributor to triggering thermal runaway by fuelling exothermic reactions under abuse conditions. The development of flame-retardant or self-extinguishing electrolyte systems is essential to mitigate thermal risks and ensure the safe deployment of lithium-ion batteries in future energy infrastructures.

Aside from material design, the evaluation of electrolyte flammability is equally important for advancing battery safety. Although various tests are employed to assess the flammability of electrolytes, the most widely used self-extinguishing time test remains unstandardized, which is often conducted under inconsistent conditions, resulting in limited reproducibility and poor comparability across different studies. Furthermore, existing assessments tend to focus on individual components (such as electrolytes) without confirming whether these improved components enhance the overall safety of the battery. The correlation between material-level safety improvements and full-cell combustion behavior remains insufficiently explored. This disconnect not only limits accurate prediction of battery thermal safety but also impedes the rational design of next-generation fire-safe materials.

This thesis aims to advance the development of fire-safe batteries through a comprehensive approach that integrates innovative component design, standardized evaluation methods, and mechanistic understanding. It seeks to establish more rigorous and systematic protocols for electrolyte flammability assessment, elucidate the underlying working mechanisms of flame retardants within the electrolyte, and reveal the correlations between safety evaluations at the electrolyte and single-cell levels. Based on these objectives, the research focus of this thesis is outlined as follows:

(i) In Chapter 3, a novel composite solid-state electrolyte with the ceramic in polymer – polymer in ceramic hierarchical structure is introduced, which exhibits enhanced mechanical properties, good fire resistance, and better electrochemical performance compared to conventional polyethylene oxide solid-polymer electrolytes.

(ii) In Chapter 4, the key parameters of the self-extinguishing time test are systematically summarized and thoroughly investigated. Based on repeatability and reliability, burning a glassfiber separator (diameter:16 mm) with absorption of 0.1 g liquid electrolyte can be proposed as a unified method. The concept of self-extinguishing efficiency is further proposed with a new evaluation criterion. The feasibility of the new protocol is verified, and the working mechanism of flame retardants in electrolytes is elucidated.

(iii) In Chapter 5, an in-depth fire safety evaluation of pouch cells containing three representative flame-retardant electrolytes is conducted using cone calorimetry. Flame retardants that exhibit excellent performance in electrolytes fail to deliver meaningful improvements in the battery`s fire safety. Finite element simulations reveal discrepancies in combustion behavior between electrolytes and full-cell levels, indicating an indirect correlation between the intrinsic flame retardancy of the electrolytes and the overall fire resistance of the batteries.