Nanoindentation of Polyethylene/Polypropylene Tri-Layer Separator Under Dry and Wet Conditions

Polymer separators play an important role in lithium ion batteries by preventing contact between the positive and negative electrodes while allowing the ions to pass through. It is, therefore, important to study not only the chemical properties, but also mechanical properties of separators. Although mechanical properties of separators have been measured using, for example, tensile tests or dynamic mechanical analysis (DMA), local mechanical properties, especially when the separators are immersed in the electrolyte, are not well known. The local deformation behavior under compression is important since the separator is usually in contact with the porous electrode layer which has micro- and nano-meter length scale surface roughness. To study the mechanical properties in a more realistic environment, a polyethylene/polypropylene tri-layer separator (Celgard 2325) was investigated, under both dry and wet conditions, using a nanoindentation system. The indentation measurements were preformed inside of an argon filled glovebox which allowed for indentation in an organic electrolyte. A piece of separator was first saturated with several drops of electrolyte before several more drops were added to create a thin electrolyte layer in which the indenter tip remained submerged during indentation to avoid the surface tension effect on the measured load vs. displacement curves. By analyzing these curves, the elastic modulus and hardness were obtained. A t-test for two population means was conducted to ascertain whether there is a statistical significance between the dry and wet conditions. Under wet conditions, the separator displayed a larger elastic modulus of 0.884 GPa and hardness of 0.0478 GPa compared to 0.801 GPa and 0.0418 GPa under dry conditions, respectively. The incompressibility of the liquid electrolyte may be responsible for the measured differences. This study provides a more representative set of mechanical properties that should be considered especially for use in computational modeling and design of battery cells.