Ionic Liquid Containing Solid Polymer Electrolytes for Lithium Metal Batteries

Researching novel battery systems is essential for the turn away from fossil fuels and electrification of cars, since renewable energies (e.g. light, wind) are not available at all times and the energy is not necessarily consumed at the place where it is gained.

Lithium air / lithium oxygen batteries state a possible candidate facing the necessity for high energy battery systems, due to a high theoretical energy density (11.680 Wh∙kg^-1), comparable with those of gasoline (13.000 Wh∙kg^-1).^[1]

Several reasons hindered the commercialization of secondary lithium air / lithium oxygen batteries so far. On the cathode side the oxygen evolution and the oxygen reduction reaction are not yet efficient without (expensive) catalysts (e.g. gold, platinum).

Like in normal lithium metal batteries, most of the challenges are related to the negative electrode material. Lithium metal is unstable against various materials, including the commonly used organic carbonate based electrolytes. While reacting with the electrolyte, the capacity of the battery is dramatically fading, and a highly resistive layer, the Solid Electrolyte Interface (SEI) is growing. The reaction of electrode and electrolyte does not only lead to solid components, but also to gaseous products, which in closed systems cause dangerous pressure increase. Within a lithium air battery also reactions with other components of the ambient air (e.g. carbon dioxide, nitrogen, moisture) have to be taken into account. During these reactions inactive materials like lithium carbonate and -nitrate are formed, disabling the cyclability very fast.

The most important reason hindering the commercialization of secondary lithium metal batteries is the formation of high surface area lithium during cycling, categorized by the morphology subdivided into dendritic, mossy and granular lithium. Dendritic lithium is the most dangerous one. The best case scenario would be a loss of contact of the dendrites to the lithium metal electrode, just leading to a fade in capacity. The worst case would be a dendrite growing through the separator to the positive electrode, causing thermal runaway and explosion of the cell. This can be prevented by the use of a solid electrolyte.

Solid electrolytes have negligible vapor pressure, are not flammable and are leakage-free. They also combine the benefits of a separator with an electrolyte, thus saving costs. Depending on the system they have low gas and water diffusion rates. One drawback compared to liquid electrolytes is the usually much lower specific conductivity.

Within the last years ionic liquids have been investigated, and they show a suppression of dendrite growth.^[2] They have higher specific conductivities than solid electrolytes, but still there are some drawbacks in open systems (e.g. in lithium air batteries) like leakage and cost. Solid polymer- and gel electrolytes offer good compromises.

In this work we show the combnation of a solid polymer electrolyte with an ionic liquid, combining the benefits of both systems. We will introduce tetrafluoroborate based Solid Polymer Electrolytes (SPEs). Compared to systems based on trifluoromethylsulfonylimide,^[3] they combine good conductivity in the desired temperature region (20-60 °C) with high electrochemical stability, while tetrafluoroborateas anion offers water stability and is cheap in production.

Free standing films with thicknesses of 80 - 100 µm were achieved (Figure 1), while keeping the mechanical properties needed for handcrafted cell building. The SPEs were characterized in respect to their electrochemical behavior in lithium metal half cells (plating/stripping), conductivities (electrochemical impedance spectroscopy - Figure 2) as well as electrochemical stability (linear sweep voltammetry) and compared to our previous prepared bis(oxalate)borate based SPEs.^[4] Also lithium metal full cells (Lithium vs. Lithium iron phosphate) were tested (cyclic voltammetry - Figure 3). The tests were carried out temperature dependent.

[1] G. Girishkumar, B. McCloskey, A.C. Luntz, S. Swanson, W. Wilcke, J. Phys. Chem. Lett., 1, (2010), 2193-2203.

[2] G.B. Appetecci, G.T. Kim, M. Montanino, M. Carewska, R. Marcilla, D. Mecerreyes, I. De Meatza, J. Power Sources, 195, (2010), 3668-3675.

[3] G.T. Kim, G.B. Appetecci, M. Carewska, M. Joost, A. Balducci, M. Winter, S. Passerini, J. Power Sources, 195, (2010), 6130-6137.

[4] R. Jakelski, M. Winter, P. Bieker, Conference Poster No. 508, 224th ECS, (2013).