To achieve high energy densities with sufficient cycling performance in all-solid-state batteries, the fraction of active material has to be maximized while maintaining ionic and electronic conduction throughout the composite cathode. It is well known that low-surface-area carbon additives added to the composite cathode enhance the rate capability; however, at the same time, they can lead to rapid decomposition of the solid electrolyte in thiophosphate-based cells. Thus, the fraction of such conductive additives has to be well balanced. Within this study we determined the electronic percolation threshold of a conducting matrix consisting of Li₆PS₅Cl and C65. Furthermore, we systematically investigated the microstructure and effective conductivity (σ_eff) of the conducting matrix. The percolation threshold p_c was determined as ~4 wt.-% C65, and it is suggested that below p_c, the ionic contribution is dominant, which can be seen in temperature-dependent σ_eff and blocked charge transport at low frequencies. Above p_c, the impedance of the conducting matrix becomes frequency-independent, and the ohmic law applies. Thus, the conducting matrix in ASSB can be regarded as an electronic and ionic conducting phase between active material particles. Additionally, guidelines are provided to enable electronic conduction in the conducting matrix with minimal C65 content.

Li redistribution processes within Si-graphite composite (SiG) electrodes are analyzed using in situ and operando X-ray diffraction (XRD), ex situ light microscopy (LM), in situ optical microscopy of cross-sectioned full cells (CS-IOM), and 3D microstructure-resolved simulations of full cells. First, the lithiation behavior of graphite and SiG full cells (Si content 20.8 wt.-%) is analyzed. The results are used as validation of the methods (XRD, LM, CS-IOM, simulation). Second, the Li redistribution between the graphite component and Si component within SiG electrodes is investigated: By operando XRD measurements during charging in comparison with relaxed cells, a higher lithiation degree in the graphite component is found during charging compared to the relaxed state, indicating Li redistribution from graphite to Si during relaxation. The Li redistribution is directly observed by in situ and ex situ optical microscopy, where the golden LiC₆ phase disappears during a 24 h relaxation period. The results are supported by simulations showing the variation in the Li concentration, not only in graphite but also within the Si component. Furthermore, all methods find that the Li redistribution is more pronounced at a higher C-rate of 0.5 C, suggesting a preference for graphite lithiation over Si lithiation.

Lithium deposition on anode surfaces can lead to fast capacity degradation and decreased safety properties of Li-ion cells. To avoid the critical aging mechanism of lithium deposition, its detection is essential. We present workflows for the efficient detection of Li deposition on electrode and cell level. The workflows are based on a variety of complementary advanced physico-chemical methods which were validated against each other for both graphite and graphite/Si electrodes: Electrochemical analysis, scanning electron microscopy, glow discharge-optical emission spectroscopy and neutron depth profiling, ex situ optical microscopy, in situ optical microscopy of cross-sectioned full cells, measurements in 3-electrode full cells, as well as 3D microstructurally resolved simulations. General considerations for workflows for analysis of battery cells and materials are discussed. The efficiency can be increased by parallel or serial execution of methods, stop criteria, and design of experiments planning. An important point in case of investigation of Li depositions are rest times during which Li can re-intercalate into the anode or react with electrolyte. Three workflows are presented to solve the questions on the occurrence of lithium deposition in an aged cell, the positions of lithium deposition in a cell, and operating conditions which avoid lithium depositions in a cell.

In this study, we investigate the use of artificial neural networks as a potentially efficient method to determine the rate capability of electrodes for lithium-ion batteries with different porosities. The performance of a lithium-ion battery is, to a large extent, determined by the microstructure (i.e., layer thickness and porosity) of its electrodes. Tailoring the microstructure to a specific application is a crucial process in battery development. However, unravelling the complex correlations between microstructure and rate performance using either experiments or simulations is time-consuming and costly. Our approach provides a swift method for predicting the rate capability of battery electrodes by using machine learning on microstructural images of electrode cross-sections. We train multiple models in order to predict the specific capacity based on the batteries’ microstructure and investigate the decisive parts of the microstructure through the use of explainable artificial intelligence (XAI) methods. Our study shows that even comparably small neural network architectures are capable of providing state-of-the-art prediction results. In addition to this, our XAI studies demonstrate that the models are using understandable human features while ignoring present artefacts.