Owing to its huge theoretical capacities, the incorporation of silicon into graphite electrodes has emerged as one of the most encouraging strategies to improve the and energy density of lithium-ion battery anodes. However, both silcon and graphite differ with the way they interact with lithium during cycling complicating the practical implementation of both Si (at high contents) and graphite in an electrode. It is well known that the role of the binder is a critical factor to the implementation of electrodes containing Li-alloying materials, such as Si, which expand and contract upon lithiation and delithiation. These volume changes are often associated with poor solid electrolyte interphase (SEI) stability, fracturing of active materials and comcomitant capacity losses. Therefore, it is of high importance to understand the binder requirements over a range of Si-graphite ratios - where what works best at high graphite content may not be sufficient enough to surpress the pressures imposed by the Si, when featured in higher content.

The focus of this study concerns the use of different aqueous-based binders: carboxy methylcellulose (CMC), polyvinyl alcohol (PVA) and polyacrylic acid (PAA) for the preparation of silicon-graphite composites anodes (at different Si : graphite ratios). Here, we show how the choice of binders affects the electrochemical performance of the silicon graphite electrodes. Among the chosen binders, the most promising electrochemical performance was achieved utilizing the PAA binder. At 15 % Si content, the best performance was observed for Si-G anodes prepared using PAA, which retained 90 % of its initial discharge capacity after 200 cycles. In second place, the next best performance was observed for anodes prepared using CMC, with the PVA binder falling into last place (all at a 15 % Si content).

Complimentary X-ray photoelectron spectroscopy (XPS) analysis, on post-cycled electrodes, suggested more effective SEI formation when using CMC and PAA binders versus PVA. The samples prepared using the PVA binder featured a more intense graphite peak which could be associated either with a thinner or more fractured SEI which is therefore unable to protect the active material upon subsequent cycles resulting in a greater rate of capacity loss. This study shows how binder optimization will be critical in order to realise the targeted energy density required by the lithium-ion battery anodes of the future.