Internal multi-dimensional signals sensing enables early warning of thermal runaway for NCM523-graphite batteries

Operando monitoring internal multi-dimensional signals of lithium-ion batteries (LIBs) during operation is critical to determine the battery state, especially, showing potential to facilitate advances in early warning of safety-related failure. However, it remains challenging to capture reliable early indicators for the early warning associated with thermal runaway. In this work, the pouch cells with the optical fiber sensor combined with the reference electrode for internal temperature and anode potential monitoring are developed, which offers non-destructive sensing techniques to unlock a trove of electrochemical and thermal data that can revolutionize the battery management system. By conducting thermal abuse tests on eight pouch cells, including the benchmark cell and the cell implanted with sensors, a two-level early warning strategy based on the battery's internal temperature (60 °C) and the decline rate of anode potential measurements (2.0 mV/min) was established. The failure mechanism related to thermal runaway can be evaluated in advance from the internal state of the battery. This conclusion is supported by thermal abuse tests on card cells, with a morphological examination carried out afterward for further analysis. Meanwhile, the results regarding the non-destructive implantation of multiple sensors were also verified in long-cycle tests (over 200 cycles at 1C) of the four pouch cells. The thermal runaway early warning strategy is proposed to against safety hazards with reproducible tests for the LIBs (NCM523-graphite), including thermal abuse tests and overcharge tests, the proposed technology provided an early warning time of more than 14 min during testing of multiple pouch cell samples. In summary, this work paves the way to commercial non-destructive sensing for advanced battery management, and it provides crucial capabilities for practical thermal runaway early warning.