Skip to main content

Scientists quantify thermal runaway propagation within Lithium-ion batteries


Lithium-ion batteries are widely used in everyday appliances, but certifying their safety in a range of applications is a key challenge. Now a team led by University College London has come up with an effective way to quantify thermal runaway within batteries using high-speed synchrotron-radiography. The results are published in Energy and Environmental Science.

  • Share

Lithium-ion batteries are a convenient power source for multiple applications including transportation and grid energy storage, contributing to next zero goals.

Their uptake is constantly growing, with currently tens of billions of batteries entering the market each year. Whilst catastrophic battery failure is very rare, and is estimated to affect around 1 in 10 million units, understanding and mitigating failure is critical.

Battery failure can happen due to electric short-circuits, overheating or through impact or penetration. This can ultimately lead to explosions, fires, and the release of toxic and flammable gases through a process known as ‘thermal runaway’. "The risk, whilst rare, is significant", explains Paul Shearing, Professor of chemical engineering at University College London (UCL) and corresponding author of the study. "We need to better understand failure so that we can engineer safer batteries, with improved solutions and mitigation strategies for failure - the understanding of the science of safety of a core part of our work with ESRF and of The Faraday Institution Safebatt project” he adds.

For the last 10 years, Shearing and collaborators at the National Renewable Energy Laboratory (NREL) in the United States and the ESRF have been developing improved battery safety testing methods using the ESRF’s beamline ID19. Now, they have come up with a toolbox to assess and quantify the safety features of batteries. "We can now quantify, for the first time, the rate of propagation of battery failure mechanisms revealed by high-speed X-ray radiography at the ESRF", explains Shearing. The team uses a unique combination of Gabor filtering and cross-correlation, which enables the toolbox to track selectively the internal structure at the onset of failure and throughout the process of thermal runaway.

The toolbox focuses on how the electrodes displace when there is nail penetration, but it can be used to analyse other failure mechanisms as well. "The true novelty of this technique is the amount of information that we can extract related to thermal runaway propagating axially and radially within a Li-ion cells, and condensing it to a single 2Dmap", explains Anand Pallipurath Radhakrishnan, Innovation Scholars Fellow from UCL. Through this method, the team noticed that thermal runaway propagation within batteries by nail penetration occurs more slowly than previously thought, which may have safety consequences when compared to other types of failure that stem from instantaneous short-circuiting events or thermally induced degradation events. For example, the rate of thermal runaway propagation and completion within a cell will determine the rate of heat emitted, the maximum temperature reached, and the necessary thermal management systems within electric vehicles to safely handle such conditions.

"The ESRF's high brilliance and high-resolution scans have been crucial to the design of the toolbox", says Shearing. “The new Extremely Brilliant Source will enable us to visualize and quantify the short-lived events that occur during battery failure with detail previously unachievable anywhere else in the world”. Alexander Rack, scientist in charge of ID19, adds: “Beamline ID19 is uniquely equipped to conduct this type of work and the team at UCL, NREL, and NASA have established a safe work environment for successfully capturing thermal runaway of batteries with high-speed X-ray imaging”. 

This tool is not only validating existing theoretical mechanical and thermal models of thermal runaway within Li-ion batteries, but also standardises battery failure testing procedures. In this way, manufacturers can track where and how failure starts in the battery. "Ultimately, if battery manufacturers start testing their batteries in this way, we are convinced that they would find strategies to improve durability and safety", concludes Mark Buckwell, battery safety researcher at UCL. Previous work from this group published in Advanced Science applied high-speed X-ray imaging to compare how the design of safety vents on commercial Li-ion cells affected the risk of the cell violently bursting and concluded that cells would be safer with a second vent to further relieve pressure. Since then, several commercial cell manufactures have adopted cell designs with secondary vents, demonstrating a clear contribution to research and development of battery safety systems. For the work published here, having demonstrated that researchers can now compare the rate that thermal runaway propagates within cells, we can now explore how factors like state of charge, cycle history, cell composition, and the use of safety devices affect the rate of propagation and therefore the safety of cells. This adds a new and impactful diagnostic tool for manufacturers and engineers to identify effective strategies for improving the safety of cells. Furthermore, to facilitate benchmarking of cell behaviours against previous cell designs, the team consisting of UCL, NREL, and NASA have and will continue to make the radiography and thermal data collected open source via the Battery Failure Databank, which is now the largest open-source repository of battery failure data in the world. This open-source approach for valuable synchrotron data is expected to provide insight to various battery manufacturers on the dynamic behavior of their cells during thermal runaway and accelerate the evolution of safe battery systems for electric vehicles.


Radhakrishnan A.N.P., et al, Energy Environ. Sci., 2022, Accepted Manuscript. DOI



Top image: A nail puncturing a cell during the experiment. Credits: Radhakrishnan A.N.P., et al, Energy Environ. Sci., 2022, Accepted Manuscript.