Abstract

Silicon has been investigated for use as a next-generation, high-capacity anode material as its theoretical lithium capacity of approximately 4140 mAhg 1 (ca. Li4.4Si) is eleven times higher than the capacity of graphite (ca. 372 mAhg ), which is currently used as an anode material. In spite of the high capacity of silicon, severe particle pulverization can be triggered by a large volume change (> 300%) during lithium alloying (to form LixSi) and de-alloying (to reform Si), which results in electrically disconnected smaller particles. These disconnected particles cause a rapid decrease in cycling stability. Intense studies have focused on reducing this volume change by using composites with an inactive carbon phase to prevent the aggregation of particle growth and to act as electrically connecting media between anode particles and the current collector when the particle is pulverized. However, these methods lead to a decrease in the charge capacity to less than 1500 mAhg 1 after dozens of cycles. On the other hand, control of the volume change by control of the morphology of the Si has very rarely been reported. Chan et al. have reported Si nanowires that showed a reversible capacity of approximately 2900 mAhg 1 at a rate of 0.05 C, which were grown on a metallic current collector. However, the capacity retention at a 2 C rate was less than 50% of the initial capacity. Ma et al. reported a first-charge capacity of 3952 mAhg 1 for nestlike Si particles, but the capacity retention of the particles was 36% between 1.6 V and 0.02 V at a rate of 0.5 C after 50 cycles. Recently, Liu and co-workers demonstrated that 3Dmetal foam structures of Cu and Sn fabricated by using an electrochemical deposition process exhibited not only fast transport of lithium ions through the electrolyte and the electrode, but also rapid electrochemical reactions, which resulted in a high performance anode with a superior rate capability. For instance, a Cu6Sn5 alloy showed a 45% capacity retention at a 20 C cycling rate, but, because of a very thick pore wall (> 100 mm), capacity fade was pronounced after 40 cycles. To date, there have been no reports of the synthesis of 3D porous Si particles, with the exception of those from the magnesiothermic reduction method. Using this method, threedimensional silica microassemblies were formed into microporous silicon replicas in a sealed steel ampoule at 650 8C by the following reaction: 2Mg + SiO2 (s)!2MgO (s) + Si (s). Herein, we report a versatile synthetic method for the formation of 3D porous bulk Si particles by the thermal annealing and etching of physical composites obtained from butyl-capped Si gels and SiO2 nanoparticles at 900 8C under an Ar atmosphere. Complete etching of the SiO2 from the SiO2/carbon-coated Si (c-Si) composite results in the retention of the remaining c-Si as a highly porous but interconnected structure, thus preserving the starting morphology. A thin pore-wall size of approximately 40 nm can accommodate large strains without pulverization, even after 100 cycles, and a maintained charge capacity of greater than 2800 mAhg 1 at a rate of 1 C (= 2000 mAhg ). SEM images of the SiO2/c-Si composites etched in HF (1m) for 2 h show that the Si particles retained their threedimensional morphology and show that the Si particles have many voids, like an “octopus foot” (Figure 1a–d). Because

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call