Due to their theoretically higher specific energy (energy/mass) and energy density (energy/volume) future electrochemical energy storage systems can provide significantly higher runtimes and ranges in the mobility sector, both in the area of mobile devices (cell phones, mobile computers, power tools, etc.) and on a large scale in electromobility.
Batteries based on metallic lithium represent a promising approach that could enable the required high specific energies and energy densities.
The use of metallic lithium in electrochemical energy storage issues new challenges for the battery community from the material to the cell to the system level. The highly reactive metallic lithium requires the formation of a protective layer on the surface of the lithium anode via a reaction of the lithium with the electrolyte (solid electrolyte interphase - SEI). While in the lithium-ion battery a solid electrode matrix, in which the lithium is incorporated, mechanically stabilizes the SEI, the purely metallic anode is built up and broken down during charging and discharging, so that the SEI is subjected to much greater mechanical stress and a change in volume. The resulting rupture of the SEI causes further contact of the lithium with the electrolyte and thus thickening of the SEI. In this process, not only are the active material lithium and the electrolyte consumed, but the transport of lithium ions between the electrodes is increasingly hampered by the thickening of the SEI with high resistance. Furthermore, the metallic lithium at the SEI grows in dendritic form. These tree-like structures pose a potential danger of breaking through the SEI and the separator and causing a short circuit through contact with the counter electrode. Schematically, the dendrite growth is shown in Figure 1.
The altered battery state due to the thickened SEI and the hazard potential due to dendrite growth must be considered at the system level. This is done via so-called battery management systems (BMS), which are used to monitor and control complex battery packs as found in electric vehicles. In this case, the states of the individual cells are to be determined and differences due to production and age are to be compensated. In addition to the control function of the BMS, defects must be detected and damaged cells must be removed from the battery pack to prevent major safety risks such as fires caused by the reactive lithium and flammable electrolytes.
The experimental characterization of the lithium surface and dendrites is significantly limited by their high reactivity with oxygen, nitrogen, water, and other substances and the low interaction with X-rays. Therefore, the metallic lithium electrode is still insufficiently characterized and the knowledge of the structure and growth rates of the layer between electrolyte and lithium is not sufficient for modelling. Most of the analysis techniques take place outside the operating conditions (e. g. ultra-high vacuum or dried) and are not informative for the processes inside the cell during electrochemical charge and discharge.
The physical and electrochemical analysis methods will be coupled in-operando as far as possible by a custom-built electrochemical cell, so that the data are significantly closer to the application than in post-mortem analyses. The developed model will be verified and improved by measurements on electrochemically symmetric Li/Li cells, using electrochemical impedance spectroscopy (EIS) and potentiostatic as well as galvanostatic cyclization.
The BMS is based on complex algorithms that extrapolate the state of a battery from the macroscopic measured quantities of the battery (current, voltage, temperature, etc.). These algorithms must be trained and verified, in particular with regard to the detection of defective and thus dangerous cells. This can be done using real batteries or battery models that simulate a particular battery state and pass it on to the BMS via a hardware component (Hardware-in-the-Loop – HiL). Real measurement data under different battery operating conditions (aged, cycled, defective, heat, cold, deep discharge, overcharge, etc.) are time-consuming and costly, sometimes dangerous, and have limited reproducibility. Using models and the HiL principle, such measurements can be avoided, however, modelling requires a detailed understanding of the processes taking place in the battery.
The planned model for metallic lithium surfaces includes dendrite growth in two dimensions, i.e. growth across the electrode surface perpendicular to the growth direction. At the same time, the growth will be combined with the (inhomogeneous) SEI layer and its influence (morphology, geometry and resistance) on the dynamics of the growth. For this purpose, SEI growth on metallic lithium surfaces will be modeled for the first time. The transport processes will be implemented according to the established Newman models, which have already been successfully developed and implemented at Fraunhofer IEE for the lithium-ion- and for the lead-acid battery. This software serves as the basis for the dendrite and SEI growth processes to be modeled. For a successful implementation of the model and for reliable results, the empirical results on both growth processes as well as the parameters of the electrochemical storage obtained from experimental measurements are crucial for the software to be developed.
Fraunhofer Institute for Energy Economics and Energy System Technology![Figure 1: Dendrite growth using a lithium-ion battery cell as an example [SLAC 2017]](/en/research_projects/search/2021/metalit/jcr:content/contentPar/sectioncomponent/sectionParsys/textwithinlinedimage/imageComponent2/image.img.4col.large.jpg/1627630751316/metalit-dendriten-wachstum-lithium-ionen-batteriezelle.jpg "Figure 1: Dendrite growth using a lithium-ion battery cell as an example [SLAC 2017]")
