Silicon has emerged as the most promising anode candidate for next-generation lithium-ion batteries (LIBs) owing to its high theoretical specific capacity of 3579 mAh/g combined with a relatively low working potential and its natural abundance. However, the practical application of pristine Si in anodes is seriously hindered by its low intrinsic electrical conductivity and the large volume variation during charging and discharging. The resulting mechanical stress causes rapid pulverization of the silicon, and isolation and disconnection of the active materials from the current collector. These failure events mostly cause a rapid degeneration of the Si electrodes.

In this study, Si nanoparticles (Si-NPs) were synthesized in a hot-wall reactor using monosilane as precursor. This process enables producing high purity Si NPs with a large production rate of up to 1 kg/h. Particle size, morphology, and crystallinity of the Si-NPs can be tuned towards optimized properties for battery applications by adjusting the synthesis parameters. In order to overcome the above mentioned failure scenarios of Si-NP-based anodes, Si/carbon nanocomposites such as Si-CNT, Si/graphene, Si-CNT/porous carbon, and Si-CNT/graphene composites have been developed to improve mechanical as well as electrical properties. As a highly promising product, the Si-CNT/ graphene nano-heterostructure demonstrates a high reversible initial capacity of 1665 mAh/g and very stable cycling performance over 500 cycles with a capacity decay of only 0.02% per cycle. Besides, the composite exhibits also a high-rate capability of 755 mAh/g (45.3% retention) at 10 C. These superior results imply that Si/carbon nano-heterostructures can be used for the development of high-performance lithium-ion batteries for durable and high-rate uses.