Nanocrystalline Li-Al-Mn-Si Foil as Reversible Li Host: Electronic Percolation and Electrochemical Cycling Stability

When a metallic foil (Li metal or LixAl) with initial Li inventory (LiInv) is used as the anode in lithium-ion batteries, its metallurgical damage state in the presence of organic liquid electrolyte and cycling electrochemical potential is of great interest. While LixAl foil operates at a voltage that eliminates Li-BCC dendrite, the state-of-health (SOH) of LixAl anode can still degrade quickly in full-cell cycling. To analyze the causes, we decompose SOH = SOHe x SOHi x LiInv, where SOHe is SOH of electronic percolation within the anode, SOHi is SOH of Li percolation from cathode to the anode interior, and LiInv is the amount of cyclable lithium in a full cell, all normalized such that 1 means perfectly healthy, and 0 means dead. Any of the three (SOHe, SOHi, LiInv) dropping to zero would mean death of the full cell. Considering the poor performance of pure Al foil due to rapid drop in LiInv, we employed a mechanical prelithiation (MP) method to make LiInv >1 initially. The chemomechanical shock from MP creates an ultrananocrystalline LiAI layer with grain size 10-30 nm on top of unreacted Al. We then monitor SOHe evolution of the anode foil by measuring the in-plane electronic conductance in situ. We find that small additions of Mn or Si into Al induce nanoprecipitates Zener pinning, and the resulting denser grain boundary (GB) network before MP significantly reduces foil porosity after MP, delays gross foil fracture, and improves SOHe in subsequent cycling. Microstructural analysis reveals that the refined grain size of foil before MP relieves stress and reduces the chance of forming electronically isolated dead grain cluster due to cracking and invasion of electrolyte and solid-electrolyte interphase (SEI). By maintaining good electronic percolation, binder-free LixAlMnSi anode demonstrates an order-of-magnitude more stable SOHe and better electrochemical cycling performance than LixAl anode.