Healing of Micro-Cracks in Silicon-Aluminum Anodes for Lithium-Ion Cells: Role of Silicon Particle Size and Electrolyte Composition

Secondary lithium-ion batteries are attractive energy storage devices with high gravimetric and volumetric capacity and the ability to deliver high rates of power. But the constantly increasing power requirements for pure and hybrid electric vehicles has challenged the existing lithium-ion battery technology, and that prompted intense interest in developing battery electrodes with still higher energy densities. As a high capacity anode and also capable of sustaining high power rates, Si has attracted attention due to its highest known theoretical charge capacity (4,200 mAhg -1 for Li 4.2 Si). Although the theoretical specific capacity of Si is more than four times higher than the Sn (991 mAhg -1 for Li 4.4 Sn) and ten times higher than the graphite (372 mAhg -1 for LiC 6 ), Si has limited applications. During electrochemical cycling in Li-ion batteries, Si anodes undergo volume changes that results in fracture and fragmentation of Si. In this work, we have considered Si electrodes as components in a Si-Al “composite” where micron-sized Si particles are embedded in a ductile Al matrix. Formation and closure of micro-cracks that caused their healing in Si particles was observed in anodes fabricated from Si-Al alloys when electrochemically cycled in ethylene carbonate (EC)- and propylene carbonate (PC)-based electrolytes (vs. Li/Li + ). In-situ optical microscopy performed during electrochemical experiments indicated the development of a network of cracks in Si particles during lithiation as the voltage decreased from 1.50 V to 0.02 V (Fig.1). Cycling in EC-based electrolyte promoted formation of a constantly growing and non-uniformly distributed electrolyte reduction products on Si (Fig.1), whereas in presence of PC, more damage of Si occurred by loss of material. Cross-sectional high-resolution transmission electron microscopy (HR-TEM) indicated that the SEI formed on Si consisted of LiCl, Li 2 O 2 and Si nano-fragments that were possibly comminuted from the Si surfaces. Optical profilometry and SEM observations were used to discuss the changes in electrode surface morphology, and FIB results revealed the evolution of Si subsurface microstructure due to cycling. The Al matrix stopped propagation of the cracks in Si. Al, having a high fracture toughness of 21 MPa.m 0.5 , exerted a larger resistance to crack advance than Si that has a fracture toughness of 1 MPa.m 0.5 (Fig.1). In-situ micro-Raman spectroscopy and HR-TEM revealed that an amorphous layer of a depth of ~100 nm formed at the surface of Si particles. Crevices were formed in this amorphous layer due to propagation of compressive shear cracks. The crevices also acted as stress concentrators to initiate the cracks in the crystalline interior. The crack surfaces also provided easy paths for Li-ion diffusion and the intercalated Li-ions formed amorphous regions ~ 5-6 nm wide on each side of the cracks. The amorphous zones having an increased volume (2.4%) applied compressive stresses causing closure of cracks as the voltage increased from 0.02 V to 4.00 V during de-lithiation. The role of the crack formation and healing on Si particle size during lithiation and de-lithiation will be discussed. It was observed that the self-healing occurred in two steps; arresting of micro-cracks at the Si/Al interfaces, and closure of the cracks at the end of the electrochemical cycle. The observation of the self-healing process should have technological implications in development of this type of composite Si particulate electrodes, as one of the major problems in monolithic Si electrodes, pulverization, could be alleviated in this way. Therefore, the results are significant for battery technology. Composite materials containing high capacity Si particles and a ductile phase around them could, therefore, provide durable electrodes for Li-ion batteries and enhance the capacity retention capability leading to prolonged battery life. Figure 1

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.