Composites of Silicon@Li4Ti5O12 and Graphite for High-Capacity Lithium-Ion Battery Anode Materials

Abstract

The application of lithium-ion batteries to long-range electric automobiles requires negative electrode materials with a higher specific capacity than traditional graphite. Next-generation materials should have both a high gravimetric capacity and capacity retention upon cycling. Silicon is a promising material for the negative electrode as it has a theoretical capacity nearly 10 times greater than graphite (3579 mAh/g for Li15Si4) (1). However, pure silicon (Si) undergoes severe mechanical stresses and pulverization during lithiation and delithiation (2). In contrast, Li4Ti5O12 (LTO) is mechanically stable and has a high rate capability. This makes it suitable to buffer the volumetric expansion of silicon when integrated into a core/shell design. In this study, a facile sol-gel technique was used to treat silicon nanoparticles and create a Si-core @ LTO-shell by modifying a procedure described previously (3). These silicon nanoparticles retain the high capacity associated with Li-Si alloys, but the pulverization of the particles is reduced. Further improvement to the cycle stability of the Si/LTO composite was achieved with the addition of graphite to the electrode material. The graphite not only provides capacity, but it is highly stable and helps buffer some of the volumetric expansion of the silicon-rich electrode material (2). The combination of graphite (G) and nano silicon/LTO (Si@LTO) is therefore an economical and scalable approach to increase the energy density of lithium-ion batteries.The electrode composites were tested in half cells and were cycled at C/10 and 1C. The cycling performance of Si@LTO + G revealed an initial high capacity (~860 mAh/g at C/10 and ~650 mAh/g at 1C) and good capacity retention (~75% after 90 cycles at 1C). In contrast, a silicon/graphite (Si/G) composite retained only 35% of its capacity at 1C under the same conditions. Compared to Si/G, the Si@LTO + G composites had lower capacity fade by reducing both silicon pulverization and SEI formation. This was confirmed by a reduction in the impedance of the Si@LTO + G batteries after cycling. Figure 1 illustrates the difference in cycle performance between Si/G and Si@LTO + G electrodes. The Si@LTO + G composite had a better rate capability when the charging rate was increased from C/10 to 1C. At 1C, the Si/G composite immediately lost 50% of its capacity. In contrast, the Si@LTO + G composite immediately lost only 25% of its capacity. This value is similar to the capacity retention of a graphite half cell tested under the same conditions (~372 mAh/g at C/10 and ~280 mAh/g at 1C). The electrode powders were characterized with X-Ray Diffraction (XRD), Scanning Electron Microscopy (SEM), Energy Dispersive X-ray Spectroscopy (EDX), and Transmission Electron Microscopy (TEM). XRD of the raw Si@LTO powder revealed the presence of crystalline silicon and Li4Ti5O12. The core-shell structure was confirmed with TEM, and EDX revealed a uniformed distribution of the silicon, titanium, and oxygen. The coin cells were characterized with operando XRD using a modified coin cell and coin cell holder designed in-house. This technique was used to observe changes in the phases of the graphite, silicon, and LTO shell. These silicon@LTO + graphite composites prove to be promising for the development of stable and high-energy-density lithium-ion batteries. Figure 1: (a) Normalized delithiation capacity vs. cycle number for both silicon/graphite and core-shell silicon@LTO with graphite. First 10 cycles at a charging rate of C/10, followed by 90 cycles at 1C. (b) Animation of a silicon particle (top) and an ideal Si@LTO particle (bottom) upon lithiation.