Lithium-ion batteries (LIBs) are playing an increasingly important role in enabling the transition to a low-carbon global economy, in electric vehicle and grid storage applications. Despite increasing LIB ubiquity, there remains gaps in our understanding of how to optimize further LIB electrode design and structure, and how to achieve such designs in practice and at scale without excessive trial-and-error experimentation. For example, thicker electrodes are desirable for improved volumetric capacity but particularly during rapid charging and discharging cycles, a significant through-electrode thickness Li ion concentration gradient in the electrolyte builds up. This leads to a spatially varying concentration overpotential that in turns leads to uneven utilization of the active material, resulting in diminished capacity and accelerated degradation. Although models of electrode dynamics can help qualitatively to understand concentration polarization effects as a function of electrode design, there are few practical experimental tools to visualize or resolve Li-ion concentration gradients in practice. For example, the widely-used energy dispersive X-ray spectroscopy (EDS) in a scanning electron microscopy cannot resolve elements with very low atomic number such as Li.In this presentation we describe the development of a methodology to visualize Li-ion concentration gradients across a range of electrodes based on secondary ion mass spectroscopy (SIMS). We combine SIMS with EDS to collect elemental maps of all principle LIB electrode elements in a single workflow, which is also extended to 3D by further combining with a plasma field ion microscope (P-FIB) capability. We apply the new methodology for LIB cathodes based on LiFePO4 (LFP) and LiMn2O4 (LMO) that are at different state of overall charge, achieved at a range charging rates. The electrodes also span a range of thicknesses from 100 to 700 µm. We show how the approach can operate over cross-sections large enough to encompass all the electrode thickness while maintaining sufficient spatial resolution to capture key features of the local Li concentration. While EDS elemental maps, with appropriate calibration, can be correlated to local element concentration with good accuracy, SIMS spectra and specifically the intensity of the 7Li+ peaks cannot be readily related to local Li concentration since the yield of 7Li+ ion relates to additional factors, such as the local atomic environment, surface topology and more. We describe how we inter-relate the EDS and SIMS maps for non-Li elements to account for some of these features, and show how these reveals the underlying Li distributions. We consider how these measurements of the Li concentration in the solid particles of the electrode relate to the local state of charge, and the conditions in the electrolyte when charge/discharge was halted and the sample was retrieved from the cell for examination. Fine-scale variations in local 7Li+ ion intensity are also revealed and explained in terms of local microstructural features such as particle size, any particle cracking, the growth of secondary electrolyte interphase, etc.
Read full abstract