Silicon-Graphite Composite Electrodes for High Energy Density Li-Ion Battery Applications

Abstract

Silicon-based electrodes are strong candidates to enable the next generation of Li-ion batteries 1. Electrochemical lithiation of silicon results in the formation of a Li-Si alloy which facilitates the uptake of significantly higher amounts of lithium ions (~3590 mAh g-1, Li15Si4) compared to conventional graphitic intercalation electrodes (372 mAh g-1) 2. In addition, silicon provides other favorable attributes, including high natural abundance, potentially low cost precursors, and non-toxicity 3. Nonetheless, commercialization of silicon electrodes is still hampered. Large volume changes up to 400% during (de-)lithiation and severe side reactions upon cycling result in large electrode polarization and capacity decay, due to loss of the electric and ionic conductivity 4. Composite electrodes, consisting of silicon and graphite active material, offer the potential to combine both high energy density with reasonable electrochemical performance. While silicon boosts the electrode specific capacity, graphite improves the cycling stability and reduces electrode degradation, e.g., by accommodating silicon volume changes, enhancing electron transport through the electrode, and increasing adhesion of the electrode coating onto the copper current collector. The electrode specific capacities of these composites easily meet the goal of ~1000 mAh g-1, which is the benchmark to achieve energy densities >350 Wh kg-1 on a cell level when combined with state-of-the-art cathode materials (150-200 mAh g-1) 5,6. In the present study, silicon-based composite electrodes with active material contents as high as 80 wt%, but different ratios of silicon (100-300 nm) and graphite particles (5-30 µm), were prepared through an aqueous ball milling procedure (Tab. 1). Equal amounts of conductive carbon fibers and lithium poly(acrylic acid) binder accounted for the remaining 20 wt% of the composite electrode. Tab. 1 Active material composition and theoretical specific capacities of the investigated composite electrodes. Active material content / wt% Theoretical specific capacity / mAh g-1 electrode Label Silicon Graphite Si:G (65:15) 65 15 2390 Si:G (50:30) 50 30 1910 Si:G (35:45) 35 45 1420 Si:G (20:60) 20 60 940 The composite electrodes were characterized in regard to their electrochemical properties by cyclic voltammetry, and their morphology by scanning electron microscopy. In addition, electrode self-discharge in fully lithiated state was investigated through electrode polarization to 10 mV and subsequent open-circuit potential measurements for >500 h. Finally, battery performance of the composite electrodes was evaluated by galvanostatic cycling in half- and full-cell setup. Fig. 1 Lithiation capacity utilization of composite electrodes in Li/Si:G coin-cells normalized to the theoretical capacities shown in Tab. 1, using 130 µL LP-57 electrolyte with 10 wt% FEC at electrode loadings of 0.9±0.2 mgactive material cm- 2. Galvanostatic currents of 300-800 mA g-1 electrode (C/3, except C/10 in the first cycle), depending on the Si:G ratio. Cut-off potentials 1.2 V and 10 mV vs. Li/Li+. Constant voltage step of 10 mV at the end of lithiation with a current limit of C/40. The data in Fig. 1 indicate substantially different cycle-life for different Si:G ratios. While all composite electrodes delivered capacities close to 100% of the respective theoretical capacity during the formation cycle at C/10, the capacity fading within the following 50 cycles increases with increasing Si:G ratio. The origin of this composition dependent degradation will be shown by examining electrode impedance and morphology as well as the extent of FEC consumption after selected number of cycles at different conditions.