The effect of cell geometry and trigger method on the risks associated with thermal runaway of lithium-ion batteries

Consideration of thermal runaway heat output variability is paramount for the development of safe lithium-ion battery assemblies. This study utilizes data gathered from fractional thermal runaway calorimetry (FTRC) ex-periments to conduct a comparative analysis of thermal runaway heat output for three cell formats (18650, 21700, and 33600) as a function of trigger method (heaters, internal short-circuiting device, and nail penetra-tion). The analysis is based on comparisons for the calculated total energy yield, fractional energy yield, heat rate, and heat flux. This study reveals that nail penetration tends to result in higher thermal runaway heat output for larger cells (21700 & 33600); these experiments also tended to result in higher fractions of the total energy being released through the cell body. The smaller cells (18650) did not appear to have significant variation in heat output as a function of trigger method. This finding suggests that, for this cell type, worst-case scenario heat output could be achievable in assembly level testing regardless of the utilized trigger method. This study also demonstrates successful translation of FTRC results, as recorded in the Battery Failure Databank, into meaningful analysis that breaks down the influence of specific conditions on thermal runaway heat output.

Automated summary

Consideration of thermal runaway heat output variability is crucial for the development of safe lithium-ion battery assemblies. This study compares the thermal runaway heat output of three different cell formats under different trigger methods and finds that nail penetration tends to result in higher heat output for larger cells, with a higher fraction of total energy being released. However, smaller cells do not show significant variation in heat output. This suggests that worst-case scenario heat output can be achieved in assembly level testing for smaller cells regardless of the trigger method used.