Silicon – with its low operating potential and much higher theoretical capacity than graphite –
is assumed to be the anode material of choice for next generation lithium-ion batteries. However,
silicon undergoes huge volume changes during lithium intercalation, deteriorating the mechanical
integrity of the electrode and resulting in a significant capacity loss during cell cycling. These issues
render the introduction of silicon-based anodes as state of the art particularly challenging. Several
research activities focused on the compensation of these volume fluctuations using different binder
types and concentrations. However, these efforts have proven only partially successful. Therefore, we
aim at a proper selection of binder system and silicon concentration offering a fair trade-off between
high capacity and low electrode degradation.
To this end, water-based silicon/graphite composite anodes with various mass ratios of silicon to
graphite were prepared. Sodium carboxymethyl-cellulose (CMC) and styrene-butadiene rubber (SBR)
were chosen as binder system, whereby CMC of differing molecular weight (90kDa, 150kDa, 1200kDa)
The adhesive strength of the electrode layers was determined using 90°-peel tests and the
micrographs of tested electrode tapes were analyzed afterward in order to determine the anode
cohesion. To evaluate the electrochemical behavior at electrode level, the prepared anodes were
paired with lithium metal, assembled to half-coin cells and cycled. Long-term cycling tests were
performed using pouch cells comprising a commercial NMC811-based cathode.
Based on the mechanical tests, the addition of silicon significantly decreases the adhesive strength of
the electrode layer to the copper current collector but it does not affect the electrode cohesion.
Moreover, the electrode adhesion shows no dependence on CMC molecular weight, indicating that
solely SBR contributes to the bond strength at the interface. In contrast to graphite-based anodes,
silicon composite anodes show no dependence of cohesive strength on the CMC Mw.
As expected, increasing silicon content results in higher specific capacity. However, this effect is
weaker at higher electrode film thickness. Results at half-cell level yield slight decrease of specific
capacity at high C-rates but almost complete recovery at lower C-rates. Cycling test results with
selected anodes paired with high capacity cathodes exhibit overall poor electrochemical performance
and the variation of the polymeric binder does not show a clear influence on the electrochemical
performance of corresponding cells.