Controlling module pressure evolution over lifetime with buffer layers: Design guidelines, prediction and validation for automotive systems

Summary:

The development of cheap, energy dense lithium-ion battery systems for electric vehicles is pushed with enormous effort. Stacking pouch-type cells in modules achieves the highest energy densities. As a consequence, however, such modules require an outer bracing to provide a medium pressure environment for optimal cell operation in terms of performance, safety and lifetime. Pouch cells develop increasing pressures in a bracing due to irreversible growth. Unfortunately, a module design tool and bracing guidelines are unavailable, partly because growth mechanisms are still disputed.

Based on a review of cell growth mechanisms suggested in literature and experimental results, we conclude the growth mechanism of a NMC111/graphite cell to correlate to loss of lithium. Based on this, we develop a novel one-dimensional mechano-electrochemical model that considers the cell stack in force equilibrium with its bracing module. It bases on module dimensions and design, stiffness data of components and cell growth values over aging as parameters. The latter is measured by observing growth of aging cells at constant pressure. The model encompasses the SOC stroke as well as pressure evolution caused by cell growth over aging. In fact, it predicts the behavior of different block designs and provides understanding as well as guidelines for proper module bracing. Specifically, we can quantify the cushioning effect of buffer layers already used in state-of-the-art automotive modules to effectively extend operation at medium pressures.

For model validation, 26 automotive modules with and without pressure pads were aged. Their pressure development is determined at several points during aging ex-situ by correlating optically measured 3D-deformations of module capping plates to a reference experiment. Such measurements of pressure development over lifetime in automotive modules have not been reported yet.

The predictions of the model are confirmed by the pressure evolution for both modules. Indeed, the significantly attenuated pressure development and extended operation at medium pressures are validated for the buffered module. Remaining errors exist at higher pressures that have to be resolved in the future. Some preliminary insights on this will be set forth. The presented pressure model is a powerful tool to quantify buffer layer and module design including the initial bracing process.