Batteries with silicon anodes: Neutron experiments show how formation of surface structures reduces amp-hour capacity
Neutrons (red arrows) detect the presence of Lithium ions which have migrated into the silicon anode. © HZB
In theory, silicon anodes could store ten times more lithium ions than graphite anodes, which have been used in commercial lithium batteries for many years. However, the amp-hour capacity of silicon anodes so far has been declining sharply with each additional charge-discharge cycle. Now an HZB team at BER II of the HZB in Berlin and the Institut Laue-Langevin in Grenoble has utilised neutron experiments to establish what happens at the surface of the silicon anode during charging and what processes reduce this capacity.
”With the neutron experiments and other measurements, we were able to observe how an inhibition or “blocking” layer forms on the silicon surface during charging that hinders the penetration of lithium ions”, explains HZB physicist Dr. Sebastian Risse. This 30-60 nanometre layer consists of organic molecules from the electrolyte liquid and inorganic components. When charging, the layer partially dissolves again so that the lithium ions can penetrate the silicon anode. However, energy is needed to dissolve the layer, which is then no longer available for storing. The physicists used the same electrolyte fluid in the experiment that is used in commercial lithium batteries.
Several cycles observed
After preliminary investigations with HZB’s BER II neutron source, the experiments at the Institut Laue-Langevin (ILL) in Grenoble provided a precise insight into the processes. ”Cold neutrons at very high flux are available at the ILL reactor. We were able to use them to non-destructively observe the silicon anode during several charge cycles”, explains Risse. Using a measuring cell developed at the HZB, physicists were able to examine the silicon anodes with neutrons during the charge-discharge cycles (in operando) and also record a number of other measurement values such as electrical resistance using impedance spectroscopy.
Efficiencies of 94 %
As soon as this inhibition layer is dissolved, the efficiency of the charge-discharge cycles increases to 94 per cent (94 % of the stored charge can be delivered again). This value is higher than that of lead-acid batteries (90 %), but slightly lower than that of batteries employing more highly developed lithium-ion technology, which deliver up to 99.9 %.
Outlook: Preventing the blocking layer
”We now want to investigate whether it is possible to prevent the formation of this inhibition or “blocking” layer by applying a very thin protective layer of metal oxide so that the capacity of silicon anodes decreases less over the course of many charge-discharge cycles”, says Risse.
The study was published in „Energy Storage Materials“: "Surface structure inhibited lithiation of crystalline silicon probed with operando neutron reflectivity". Arne Ronneburg, Marcus Trapp, Robert Cubitt, Luca Silvi, Sébastien Cap, Matthias Ballauff, Sebastian Risse.
- Magnon momentum microscopy: A new window into nanoscale spin-wavesAn international team lead by the Max Born Institute has developed a new type of momentum microscopy to image magnons — the quanta of collectively excited spins — directly in two-dimensional reciprocal space using soft X-rays. Measurements have taken place at BESSY II and PETRA III, first author ist the HZB physicist Steffen Wittrock. Owing to its remarkable sensitivity, simplicity, and access to nanometer-scale wavelengths, this novel technique establishes a powerful and versatile platform for exploring nonlinear magnon interactions, which are promising for future computing schemes.
- X-ray analysis reveals overpainted fascist symbolsErich Mercker was a successful painter during the Nazi era and in the years that followed. After 1945, he covered up Nazi symbols in at least one of his paintings. With an interdisciplinary team, physicist Dr Ioanna Mantouvalou reports on this study in the Nature Journal Heritage Science.
- AI agents deliver results – but do they reason scientifically?A research team co-led by Kevin Maik Jablonka from the Helmholtz Institute for Polymers in Energy Applications Jena (HIPOLE Jena) and N. M. Anoop Krishnan from the Indian Institute of Technology Delhi has developed Corral, a new benchmark for AI agents in science. The preprint “AI scientists produce results without reasoning scientifically” has been published on arXiv (https://doi.org/10.48550/arXiv.2604.18805). The analysis shows that current systems can execute scientific workflows and deliver results; however, they often do not follow the basic principles of scientific testing and reasoning.