New Strategies Emerge to Curb Thermal Runaway in EV Batteries
As electric vehicles (EVs) surge in global adoption, the safety of their lithium-ion batteries remains a critical concern for automakers, regulators, and consumers alike. Among the most pressing challenges is thermal runaway—a self-sustaining chain reaction within battery cells that can lead to fires or explosions. Recent research led by Chen Guohe, Lyu Peizhao, Li Menghan, and Rao Zhonghao from Hebei University of Technology provides a comprehensive analysis of how thermal runaway propagates in battery modules and evaluates cutting-edge strategies to mitigate its risks. Their findings, published in Energy Storage Science and Technology, illuminate pathways toward safer, more resilient EV battery systems.
Thermal runaway begins when internal side reactions in a lithium-ion cell—such as decomposition of the solid electrolyte interphase (SEI) layer or reactions between the cathode and electrolyte—generate heat faster than it can be dissipated. This triggers a domino effect: rising temperatures accelerate further exothermic reactions, leading to rapid temperature spikes, gas generation, and, in severe cases, cell venting or ignition. While modern EVs incorporate sophisticated battery management systems, the propagation of thermal runaway from one cell to neighboring cells in a pack can compromise an entire module, posing significant safety hazards.
The researchers emphasize that understanding the propagation mechanisms is as vital as preventing the initial event. Heat spreads through three primary pathways: conduction via cell casings and interconnects, convection from hot gases and vented materials, and radiation from flames or glowing surfaces. Depending on battery geometry—cylindrical, prismatic, or pouch—the dominant heat transfer mode varies. For instance, in tightly packed prismatic modules, conduction through the aluminum casing accounts for the majority of heat transfer to adjacent cells. In contrast, cylindrical cells spaced closely together may experience significant radiative heating, especially when flames are present.
A host of factors influence how quickly and extensively thermal runaway propagates. Battery chemistry plays a crucial role. Cells with nickel-rich cathodes like NCM811 exhibit faster and more violent propagation compared to those using lithium iron phosphate (LFP), which is inherently more thermally stable. State of charge (SOC) also matters—higher SOC means more stored chemical energy available to fuel exothermic reactions. Experiments show that a 10 Ah NCM pouch cell at 100% SOC can trigger neighboring cells in under two minutes, whereas the same cell at 20% SOC may not propagate at all.
The physical design of the battery pack further modulates risk. Connection topology—whether cells are wired in series, parallel, or hybrid configurations—affects both electrical and thermal behavior during failure. Parallel connections, while beneficial for current distribution during normal operation, can exacerbate thermal runaway by enabling current redistribution that overheats adjacent cells. One study cited in the review found that a 1S10P cylindrical module experienced full propagation within minutes of a single-cell failure, while a 10S1P configuration showed no propagation under identical conditions.
Battery arrangement also influences propagation dynamics. Vertical stacking, common in many EV platforms, allows flame jets and hot gases from a lower failed cell to directly impinge on cells above, accelerating upward propagation. In contrast, brick-style or staggered layouts can disrupt direct flame paths and increase the distance heat must travel, thereby slowing or halting cascading failures. Researchers demonstrated that such geometric interventions could effectively contain thermal runaway within a single module, preventing system-wide collapse.
Environmental conditions add another layer of complexity. At high altitudes or in aerospace applications, reduced ambient pressure alters combustion characteristics and heat transfer. Counterintuitively, low-pressure environments can either suppress or accelerate propagation, depending on cell chemistry and module configuration. For example, LFP modules showed increased propagation risk below 70 kPa, while NCM523 modules exhibited reduced propagation under the same conditions. This underscores the need for application-specific safety validation, especially as EVs expand into diverse operating environments.
Against this backdrop, the research team systematically evaluates thermal management strategies designed to interrupt propagation. Passive and active cooling methods each offer distinct advantages and trade-offs.
Air cooling, though simple and low-cost, provides limited thermal buffering during runaway events. However, forced airflow—particularly longitudinal wind—can significantly suppress flame development. One experiment found that airflow exceeding 4.5 m/s eliminated visible flames after venting, reducing secondary ignition risks. While insufficient as a standalone solution for high-energy packs, air cooling may serve as a complementary measure in lower-risk applications.
Liquid cooling systems, widely adopted in premium EVs, offer superior heat removal capacity. Indirect liquid cooling—where coolant flows through channels adjacent to cells—can delay propagation but often cannot prevent it once runaway initiates. Microchannel and serpentine cold plates enhance performance, yet require high flow rates (e.g., 96 L/h) to effectively suppress temperature rise in adjacent cells. Crucially, the speed of system response matters: increasing coolant flow immediately after detection of an anomaly can reduce peak temperatures in neighboring cells by over 100°C, buying critical time for intervention.
More promising is direct immersion cooling, wherein cells are submerged in dielectric fluids such as transformer oil, silicone oil, or engineered fluorocarbons like Novec 649. These fluids absorb heat directly from cell surfaces and suppress combustion by limiting oxygen access. In one test, a 60 Ah NCM622 pouch module submerged in Novec 649 showed surface temperatures capped at 184°C during overcharge-induced runaway—far below the 500–800°C typical in air—and exhibited no flame or propagation. Immersion systems also eliminate the need for complex cold plates, potentially simplifying pack design.
Phase change materials (PCMs) offer passive thermal buffering by absorbing large amounts of heat during melting. However, traditional organic PCMs like paraffin are flammable, posing an additional fire hazard. To address this, researchers have developed flame-retardant composites—blending PCMs with additives like aluminum hydroxide, ammonium polyphosphate, or melamine. These formulations delay thermal runaway onset by several minutes and reduce propagation speed. For instance, a PCM with a 10:6:3 mixture of ammonium phosphate, melamine, and pentaerythritol extended propagation time in a 10 Ah NCM pouch module by 90 seconds. Inorganic PCMs, such as sodium acetate trihydrate, eliminate flammability but face challenges with phase separation and supercooling, requiring encapsulation or composite structuring.
High thermal conductivity materials—graphite composites, aluminum plates, or metal foams—can redistribute heat away from hotspots, preventing localized temperature spikes. Yet in large, high-energy modules, these materials may inadvertently accelerate propagation by conducting heat more efficiently to neighboring cells. Thus, their use must be carefully calibrated. Conversely, thermal insulation materials act as barriers. Aerogels, particularly silica-based variants, exhibit ultra-low thermal conductivity (<0.02 W/m·K) and stability above 600°C. A 6.9 mm aerogel sheet successfully blocked propagation in a high-nickel (NCMA) module with surface temperatures exceeding 800°C. Similarly, hollow glass microsphere (HGM) boards just 3 mm thick halted runaway in 51 Ah NCM811 prismatic cells.
Recognizing the limitations of single-method approaches, researchers are increasingly exploring hybrid systems. Combining aerogel insulation with liquid cooling, for example, leverages the insulation’s ability to delay heat transfer and the coolant’s capacity to remove energy over time. One hybrid design kept adjacent cell temperatures below 90°C during a triple-cell failure event. Another system integrated PCM, aluminum plates, and liquid flow—enhancing normal-operation temperature uniformity while providing multi-stage thermal defense during abuse.
Looking ahead, the authors identify three key research directions. First, deeper mechanistic understanding of the venting and combustion processes during runaway—particularly the role of solid particulates and aerosolized electrolytes—is needed to refine propagation models. Second, multi-scale, multiphase simulation tools that couple electrochemical, thermal, fluid dynamic, and combustion physics must be developed and validated, with reduced-order models enabling faster design iteration. Third, next-generation thermal management systems should unify normal-operation efficiency and emergency suppression capabilities, potentially through smart materials that activate only under extreme conditions.
For the automotive industry, these insights carry immediate relevance. As automakers push toward higher energy densities to extend range, safety cannot be an afterthought. Battery architectures must integrate thermal propagation barriers from the outset—not as add-ons, but as core design elements. Regulatory standards, such as China’s GB 38031–2020, already mandate thermal runaway propagation resistance, and similar requirements are emerging globally. The strategies reviewed here provide a roadmap for compliance without sacrificing performance.
Ultimately, the goal is not just to contain failure, but to prevent it. While no system can guarantee absolute safety, layered defenses—combining robust cell chemistry, intelligent pack architecture, responsive thermal management, and advanced monitoring—can reduce the likelihood and consequences of thermal runaway to negligible levels. As EVs become mainstream, such engineering rigor will be essential to maintaining consumer trust and enabling sustainable mobility.
Author Information: Chen Guohe, Lyu Peizhao, Li Menghan, Rao Zhonghao, School of Energy and Environmental Engineering, Hebei University of Technology, Tianjin 300401, China.
Journal: Energy Storage Science and Technology
DOI: 10.19799/j.cnki.2095-4239.2024.0091