Silicon shows promise as a next generation lithium-ion battery anode due to its high theoretical capacity (3579 mAh/g). Volume expansion during cycling, subsequent fracturing and pulverisation of the coating, and continuous solid-electrolyte interface (SEI) formation, limit its commercial viability. Investigated solutions to improve stability on the material level include decreasing the Si particle size to the nano range, to mitigate the cracking issue[1], and compositing/alloying Si to obtain a conversion type battery chemistry, which form a robust SEI[2].

Gas-phase synthesis is a versatile and scalable bottom-up approach to produce high purity nanoparticles with precise control over their size and composition. Here, we use a hot-wall reactor to synthesize silicon-rich silicon nitride (SiNx) nanoparticle powder via the pyrolysis of monosilane (SiH4) and ammonia (NH3) gases. Production rate as high as 1 Kg/hr eases the process scale up to industry scale. Process-property correlations within this reactor design are explored with elemental analysis, XRD, BET, and TEM. The powders are further processed for electrochemical testing in liquid electrolyte and solid-state cells.

Elemental analysis shows that nitrogen content in SiNx increases with NH3 concentration and holds a linear relation at high concentrations. The specific surface area can be analysed using BET from which we can estimate particle sizes from 55 nm to 250 nm depending on the reaction parameters. These values are further confirmed in TEM.
The identification and quantification of any crystalline phases can be done using TEM and XRD. A decrease in crystallinity with increasing nitrogen content is observed.

Battery tests show a low coulombic efficiency (CE) in very small particle size powders. Higher nitrogen content in SiN yields better cyclic stability, yet at reduced capacity and first cycle CE. Formation of the c-Li15Si4 phase, which reacts with the electrolyte, is suppressed with increasing N
content (x) in SiNx. The liquid electrolyte cells with highest x (SiN0.81) retains 1000 mAh/g capacity with CE > 99.5% over 150 cycles.
The stoichiometry of SiNx can be tuned in the full range of x = 0 (pure silicon) to x = 1.33 (stoichiometric Si3N4), while typically the core of the particles is slightly more silicon rich. At specific synthesis conditions, core shell structures can be obtained.

Low surface-to-volume ratio and reactivity of large particles result in better CE during formation cycles. We can infer that high nitrogen content and amorphous nature is conducive to the cyclic stability of the anode.

[1] X. H. Liu, L. Zhong, S. Huang, S. X. Mao, T. Zhu, J. Y. Huang, ACS Nano 2012, 6, 1522.
[2] Z. Liu, Q. Yu, Y. Zhao, R. He, M. Xu, S. Feng, S. Li, L. Zhou, L. Mai, Chem. Soc. Rev. 2019, 48, 285.