A method is presented discussing how to reliably and quantitatively detect leakage from battery cells through the detection of escaping liquid electrolyte vapors, typically dimethyl carbonate (DMC). The proposed method does not require the introduction of an additional test gas into battery cells. The test system, which is non-destructive in nature, is applicable to non-rigid pouch cells and rigid prismatic or cylindrical cells.
Lithium-ion batteries are a more suitable energy source for many applications because of their high energy density and low self-discharge rate. In the automotive powertrain sector, the lithium-ion battery market share is growing rapidly, with particularly high demand being placed on battery service life and safety. Requirements regarding maximum cell temperature, electrical load power or discharge power of the cell can be controlled by cooling and power management of the battery cell. A single defect in a cell housing can only be detected through leak detection of each battery cell. The lifetime of a battery strongly depends on the tightness of the cell housing, because of the harmful effects caused by the interaction between the electrolyte and water vapor. Rapid detection of leaks in the production of battery cells is absolutely essential to achieving necessary service life and safety requirements. This applies particularly to small leaks that cannot be detected immediately after the cell has been manufactured, for example by using an electrical discharge method. For pouch cells, no reliable method to detect small leak channels is available. This paper examines the spectrum of possible leak scenarios for cylindrical, prismatic and pouch lithium-ion batteries [Figure 1]. Currently no rejection limits have been codified for these batteries. INFICON has established empirically derived rejection limits that will be discussed in this paper.