Silicon is a promising anode material for increasing energy density in lithium-ion batteries. In composite with graphite, the present dominant anode material, it allows to increase progressively the capacity of the anode by up to a factor of 10, and the overall energy density of the battery by up to 30%. However silicon has to be incorporated as a nanomaterial in the electrode to avoid fast decay due to pulverization while cycling. I will briefly review the forms of nano-silicon for lithium battery anodes, the challenges in synthesis and formulation, and the effect on performance in cycling. We recently grew a series of silicon nanoparticles and nanowires of different sizes, and cycled them in the same conditions [1]. I will present the result of this comparison, showing unexpected effects of size and shape of silicon on its electrochemical behavior in lithium batteries. Our team proposes a simple process of direct growth of silicon nanowires (SiNW) on graphite, containing up to 40wt.% Si. The interest of this compact, robust process will be discussed. This micropowder can directly enter in the slurry for the preparation of lithium battery anodes [2] following standard procedures. This material can be optimized for a specific capacity of 1000mAh/g, and holds more than 500 cycles at 2C (30 minute charge, 30 minute discharge). A large irreversible capacity loss is observed in the first cycle, as for many nanomaterials, but then the specific capacity retention is unusually high for silicon-rich anodes, and they can be incorporated in lithium-ion batteries vs NMC cathodes. I will finally address material-design issues to increase the performance of the material. In particular, I will discuss replacing gold by tin as the metallic catalysts required for SiNW growth, and possible means to reduce the irreversible capacity loss in the first cycle. [1] C. Keller et al., Nanomaterials 2021, 11, 307, 10.3390/nano11020307 [2] S. Karuppiah et al., ACS Nano 2020, 14, 12006, 10.1021/acsnano.0c05198