2, Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States
Lithium-ion batteries (LIBs) are highly important to mobile electronics, green-energy storage, and electric-vehicle technology. Commercial LIBs use relatively low-capacity graphite for anodes (372 mAh g-1). In contrast, the theoretical capacities of alloying anodes for LIBs are much higher (4200, 1600 and 990 mAh g-1 for Si, Ge and Sn, respectively). The drawback to using Si, Ge and Sn in anode construction is the roughly 300% volume change during the cycling process, which can lead to pulverization, unstable solid-electrolyte interphase, and rapid capacity loss. Herein we report the use of carbonized asymmetric membrane electrodes that contain silicon, germanium, tin dioxide or vanadium (V) oxide can significantly increase LIB electrode capacity while maintaining a long cycling life and an excellent rate performance. It is the first time that asymmetric membranes are proposed to be employed as lithium ion battery electrodes. These asymmetric membrane electrodes were fabricated using a novel adapted reverse-osmosis membrane technology, the phase-inversion method. The asymmetric membrane electrodes have a thin, nanoporous top layer (up to several µm) and a thick, macroporous bottom layer (100-200 µm). The top layer can prevent the leaching of fractured anodes, and the bottom layer provides solid mechanical support and void space that can efficiently accommodate the large volume changes driven by lithium insertions and extractions. It was demonstrated that 90% capacity of asymmetric membrane containing Si nanoparticles can be retained after 200 cycles with an initial capacity loss of ~ 30%. We also synthesized SnO2 and V2O5 asymmetric membranes using a unique combination of phase inversion and sol-gel chemistry. The resulting SnO2 electrode demonstrated a specific capacity of 500 mAh g-1 based on the overall electrode mass at a current density of 280 mA g-1 (~0.5C) with >96% capacity retention after 400 cycles. When the current density was increased from 28 to 560 mA g-1, its overall capacity was reduced by only 36%. The V2O5 electrode can be cycled at 0.5C with a capacity of 160 mAh g-1 for 380 times with ~100% capacity retention. Additionally the same method can be extended to stabilize micron-size Ge and Si alloying anodes with impressive cyclability. Finally, this method can be scaled up using a roll-to-roll process common in the membrane industry, breaking a huge barrier to the potential commercialization of the aforementioned electrode materials.