Modern technology, from portable electronics to electric vehicles, is increasingly reliant on lithium-ion batteries for energy storage, but their use poses safety concerns, namely due to the risk of thermal runaway [1]. This process results in toxic off-gassing, smoke, fire, and explosions, causing damage to property and endangering people. Therefore, early detection of thermal runaway is critical. Voltage, current, and temperature sensors are used to monitor batteries and identify the onset of thermal runaway, but these have limitations. For example, the heat generated inside the battery will take time to be detected due to the low thermal conductivity of polymers and other battery components. Past studies have measured the stress generated in each electrode during cycling [2] and investigated the volumetric expansion of batteries during cycling [3,4]. Therefore, strain measurement of cells could provide insight into the reversible and irreversible phenomenon inside of a battery in real time. To this end, strain and temperature response of a commercially available high capacity (3.4 Ah) 18650 lithium-ion battery were investigated in this study. Strain measurements were conducted while cycling in a temperature-controlled environment. During battery cycling the maximum strain was recorded at 100% state-of-charge (4.2V) in the hoop direction at the center of the battery. Strain was returned to a baseline value upon discharge. In all the cycling scenarios, the electrochemical reaction peaks coincided with change in strain response. Repeated measurements across multiple cycles and multiple identical cells have produced similar strain behavior. The results from this study indicate that strain measurements on a battery’s surface are directly correlated to the reversible electrochemical behavior of a battery, and therefore be used as a diagnostics tool for battery operation to detect irreversible phenomenon. This presentation will describe the methods of strain measurement as well as potential applications of this technique as a diagnostic or early fault detection tool.

Figure 1: Middle hoop strain and battery potential during a full cycle.