It is demonstrated that energy-filtered transmission electron microscope enables following of in situ changes of the Ca-L_2,3 edge which can originate from variations in both local symmetry and bond lengths. Low accelerating voltages of 20 and 40 kV slow down radiation damage effects and enable study of the start and finish of phase transformations. We observed electron beam-induced phase transformation of single crystalline calcite (CaCO₃) to polycrystalline calcium oxide (CaO) which occurs in different stages. The coordination of Ca in calcite is close to an octahedral one streched along the direction. Changes during phase transformation to an octahedral coordination of Ca in CaO go along with a bond length increase by 5 pm, where oxygen is preserved as a binding partner. Electron loss near-edge structure of the Ca-L_2,3 edge show four separated peaks, which all shift toward lower energies during phase transformation at the same time the energy level splitting increases. We suggest that these changes can be mainly addressed to the change of the bond length on the order of picometers. An important pre-condition for such studies is stability of the energy drift in the range of meV over at least 1 h, which is achieved with the sub-Ångström low-voltage transmission electron microscope I prototype microscope.

Energy storage is one of the major challenges on the road to environmental protection. The most commonly used batteries at present and in the near future are Li-ion-based systems. Therefore, reliable characterization methods are essential for further improvement of the properties and understanding of aging mechanisms. The imaging capabilities of scanning electron microscopy provide insight into morpholgy, structure and qualitative material contrast down to the nanometer scale [ 1, 2]. Examination of the qualitative chemical composition can be obtained with several additional equipment and correlative methods like Raman, mass spectrometry and energy dispersive X-ray (EDS) analysis [ 3, 4]. EDS enables also quantitative analysis into the trace constituent range (0.01 ≤ C ≤ 0.1) [ 5], but consistently upcoming questions are: How reliable are EDS studies for real life application under not ideal conditions, what are the influences of etc. sample preparation, storage and spectrum processing. Therefore a systematic study was carried out on Li(Ni_xMn_yCo_z)O₂ cathode material. The first topic was the determination of the stoichiometric composition of Ni, Mn and Co, where Ni, Mn and Co share one position, consequently x+y+z = 1. Table 1 show the different settings: Without contact to air (glove box => shuttle => SEM vacuum, shuttle) or prepared in air (air). Top view on a piece of cathode foil (topographic surface) or cross section obtained with Ar-Ion milling or mechanical polishing (Sec, flat surface). Additionally coated with Au (4nm) or C (10nm). Spectrum processing performed standard-based or standard-less and is automatically normalized to 100%, neglecting the Li content (Normalized). All performed examinations deliver a stoichiometric relationship close to NMC (811), Figure 1. This proves the robustness of this analysis technique. Additionally results obtained with ICP-OES (Inductively Coupled Plasma - Optical Emission Spectrometry) dotted lines in Fig.1 agree nicely (Ni: 0,835; Co: 0,107; Mn: 0,058) with the EDS results. Nowadays metallic Li becomes detectable by the new generation of EDS spectrometer, but a persistent limitation for chemical quantification of Li-ion battery materials is that bonded Li cannot directly detected by EDS. We determine the Li amount as the missing element fraction of not normalized quantitative EDS analysis. Additionally we solely take those spectra for the quantification into account, where the oxygen weight percent of the EDS spectra is ±3% of the expected weight percent for oxygen in Li(Ni₈Mn₁Co₁)O₂ . We obtained mean values of the Li weight percent for the analytic conditions: shuttle: 6,0±0,5%; air: 6,1±0,8%; Au coated: 10,6±0,8%; C coated: 10,4±0,7%, Argon Ion milling: 6,4±0,6%. Point and area spectra were acquired for all settings, providing consistent results.The mechanical polishing preparation alone yields 5.6±0.3% for the point spectra and 14.2 ±1.9% for the area spectra for the Li determination. On the one hand, this is probably due to the loss of Li during sample preparation => higher Li values for the area spectra and on the other hand by contamination during the acquisition of the point spectra underestimation of the Li content. The results of the studies performed with the conditions shuttle, air and Ar-Ion milling agree nicely with the expectation of 6,4% for a 100% lithiated cathode material after the first cycle, where the gold, carbon coating lead to a significant overestimation of the Li content and the mechanical polished section cannot deliver reliable results [ 6] .

We examine the impact of water (160 ± 41 ppm of reference) on the anode, cathode, separator and electrolyte in two aging scenarios: calendric aging (60 °C, 80 days, charged state), resulting in a triggered current interrupt device (CID), and cycling 1680 times (charge/discharge with 1C, 2.75–4.2 V, 20 ± 2 °C), resulting in 24.5% residual capacity. We applied computer tomography (CT), scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS) and secondary ion mass spectrometry (SIMS) to understand the capacity loss. The aged NMC/LCO–graphite cells were compared to a reference cell in pristine state. Both aging scenarios showed (a) thick depositions on the anode and cathode consisting mainly of oxygen, fluorine and phosphorous, (b) reduced separator pore sizes, (c) the deposition of Mn, Co and Ni on top of the anode and (d) the decomposition of the conductive salt LiPF₆ accompanied by HF formation and a loss of active lithium. Calendric aging consumes the water content and additionally leads to (e) the decomposition of the organic solvent followed by CO₂ gas formation. Cyclic aging increases the water content and additionally results in (f) the consumption of the additive fluoroethylene carbonate (FEC). These findings show how water affects aging phenomena and results in a capacity decrease in the cell.

Electrochemical aging can alter the mechanical abuse tolerance of lithium-ion cells, but evidence for large-format tabless cylindrical cells remains limited. This study investigates commercial steel-cased 4695 cells under quasi-static hemispherical indentation at 100% State-of-Charge. Fresh cells were compared with two aged groups brought to 80% remaining capacity by either room-temperature cycling at 20 °C and 20 A or low-temperature/high-current cycling at −10 °C and 45 A. Aging shifted the onset of a major internal short circuit to lower displacement, force, and absorbed mechanical work. Relative to fresh cells, the mean displacement at internal short-circuit onset decreased by 15.9% after aging at 20 °C and 20 A and by 22.1% after aging at −10 °C and 45 A, while the corresponding force decreased by 10.8% and 19.4%. The absorbed mechanical work to short-circuit onset decreased by 24.0% and 37.3%, respectively. Peak surface temperatures did not clearly separate the groups, whereas integrating the surface-temperature rise over the first 30 $\mathrm{s}$ after short-circuit onset revealed an increase from 6894 $\mathrm{K}$$\mathrm{s}$ in fresh cells to 9872 $\mathrm{K}$$\mathrm{s}$ and 13,777 $\mathrm{K}$$\mathrm{s}$ in the two aged groups. The experiments also revealed a change in observed venting behavior, with top-terminal venting occurring more frequently in aged cells. These results indicate that, under the tested conditions, aging history can reduce mechanical abuse tolerance and modify post-failure severity in large-format cylindrical cells.