High Carbon Doping of Amorphous Silicon Nanoparticles in a One-Step Gas-Phase Process for Increased Cycling Stability

Silicon as an anode material for Lithium-Ion-Batteries (LIB) is a promising candidate to increase overall specific energy of batteries in comparison to state-of-the-art graphite anodes. Yet the cycling stability of silicon-based electrodes is still an issue, mainly due to problems related to the swelling of silicon during lithiation to up to 300 % of its original volume. While a breakage of silicon particles can be prevented by choosing a nano-material this has the detrimental effect of increasing the surface exposed to the electrolyte, i.e. reducing coulombic efficiency. Other solutions like coatings of silicon particles and complex electrode structures are shown to increase cycling stability, but are typically connected to higher manufacturing cost. By doping silicon nanoparticles already during production in a gas-phase one-step process, this study investigates the influence of high carbon concentrations up to 20 wt.%. This causes the silicon to stay amorphous, even at elevated temperatures. Furthermore it is shown, that the disturbance of the lattice by carbon atoms helps mitigating stress during lithiation, and can form a less reactive surface. By varying precursor concentrations during material synthesis, a range of materials were produced and analyzed physicochemically and electrochemically. For carbon values up to 15 wt.% our results show a strong increase in cycling stability of up to 80 % capacity retention after 500 cycles, and first cycle coulombic efficiencies of 80 – 90 %. Yet the increased stability is going along with a decrease in specific capacities to 1000 – 1500 mAh/g. These values were obtained for a 75 wt.% silicon, 15 % PAA, 10% C65 electrode and 1 mg/cm2, tested against lithium at 0.5 C. By linking further electrochemical results to particle properties the effects of the carbon doping are further unravelled. This material engineering offers another promising pathway to increase overall performance of silicon containing LIBs and is industrially viable, due to its inclusion in a scalable one-step material synthesis process. Figure 1