Lithium metal is the most promising anode for next-generation lithium batteries. Nonetheless, its non-uniform Li nucleation, dendritic growth, and infinite volume variation lead to rapid cell failure and hinder its practical application. Here, we propose a defective graphene shell grown on a carbon matrix as a stable lithium metal host via carbide-mediated catalytic hydrogenolysis. It has numerous nano-channels with defective graphene shells on its surface, which can effectively guide the dendrite-free Li deposition and accommodate a rational amount of metallic Li without severe cell volume change. Due to these features, the host exhibits good cycle stability (87.2% after 500 cycles) and low dimension variation (9 μm) in carbonate electrolyte full-cell evaluations. In addition to its good performances in the carbonate electrolyte, our versatile Li host, acting as a Li-ion flux regulating interlayer, also achieves an improved cycling performance in the argyrodite Li₆PS₅Cl based high energy density all-solid-state battery full-cell configuration.

Silicon (Si) is considered a potential alternative anode for next-generation Li-ion batteries owing to its high theoretical capacity and abundance. However, the commercial use of Si anodes is hindered by their large volume expansion (∼ 300%). Numerous efforts have been made to address this issue. Among these efforts, Si-graphite co-utilization has attracted attention as a reasonable alternative for high-energy anodes. A comparative study of representative commercial Si-based materials, such as Si nanoparticles, Si suboxides, and Si–Graphite composites (SiGC), was conducted to characterize their overall performance in high-energy lithium-ion battery (LIB) design by incorporating conventional graphite. Nano-Si was found to exhibit poor electrochemical performance, with severe volume expansion during cycling. Si suboxide provided excellent cycling stability in a full-cell evaluation with stable volume variation after 50 cycles, but had a large irreversible capacity and remarkable volume expansion during the first cycle. SiGC displayed a good initial Coulombic efficiency and the lowest volume change in the first cycle owing to the uniformly distributed nano-Si layer on graphite; however, its long-term cycling stability was relatively poor. To complement each disadvantage of Si suboxide and SiGC, a new combination of these Si-based anodes was suggested and a reasonable improvement in overall battery performance was successfully achieved.

Full cell performance of conventional graphite and double-layer electrode of dCNT-G and G (dCNT-G/G) electrode at reverse designed N/P ratio (0.8). Inset images are schemes of each electrode at lithiated state with extra lithium layer.