Experimental Investigation of Thermal Runaway in Lithium-Ion Batteries: Implications for Tunnel Safety and Structural Integrity

The increasing use of lithium-ion batteries (LIBs) in electric vehicles and energy storage systems raises new safety concerns, particularly in confined environments such as road tunnels. This study presents an experimental investigation into thermal runaway (TR) events in LIBs triggered by mechanical, thermal, and electrical abuse conditions. The work focuses on characterising temperature and pressure evolution, gas emissions, and the effects of battery fires on surrounding structural materials, specifically concrete, representative of tunnel linings. Abuse tests, including nail penetration, overheating, and external short circuits, were conducted using an Accelerating Rate Calorimeter (ARC). The maximum temperature, pressure and released gases were monitored. Pre- and post-mortem computed tomography analyses provided insight into internal cell degradation and structural changes. Importantly, TR events in proximity to cement mortar samples revealed their thermal damage, leading to microstructure degradation (mostly microcrack network development) after repeated exposure. The results show that thermal runaway of cells can rapidly propagate in confined spaces, developing flammable gases, toxic substances and thermal loads capable of compromising tunnel structures. Cells triggered at higher ambient temperatures are more reactive, reaching higher peak temperatures and pressures during TR. The state of health (SOH) of cells influences the severity of TR, with degraded cells producing less intense but still dangerous reactions. Overheating tests demonstrated a critical delay window (~350 s) enabled by safety valve activation, crucial for mitigation in real tunnel scenarios. For external short circuit, no thermal runaway occurred for different SOC states. These findings contribute to a better understanding of LIB behaviour under critical conditions and support the development of safety strategies, emergency response plans, and structural containment systems for tunnels. The data will serve as a basis for future risk assessments and engineering models aimed at improving resilience to battery-induced fires in underground infrastructure.