An international team led by researchers at UCL has revealed new insights into the workings of a commercial Li/MnO₂ primary battery by virtually “unrolling” its coil of electrode layers using an algorithm designed for papyrus scrolls.

In an open-access study published in Nature Communications, the researchers combined X-ray and neutron tomography to track the processes deep within a lithium battery during discharge. They then used a mathematical model designed for ancient manuscripts too sensitive to be physically opened to “unroll” the electrode layers, so aiding analysis and revealing that different sections of the battery were operating differently.

Researchers found that using the two complementary imaging techniques and “unrolling” the electrodes while they are in normal use provides a fuller and more accurate understanding of how the battery works and how, where and why it degrades over time. Unseen trends in the spatial distribution of performance in the cells were observed.

The method paves the way for developing strategies for improving the design of cylindrical cells using a range of battery chemistries, including by informing better mathematical models of battery performance. As such the method may facilitate improvements in the range and lifetime of electric vehicles of the future.

The project was funded by the Faraday Institution, as part of its battery degradation project.

Reconstructed tomograms from neutron and X-ray computed tomography. Clearly visible in the X-ray images is the nickel current collecting mesh, which appears brighter than the active electrode material. Source: Faraday Institution

The team investigated the processes occurring during discharge of a cylindrical commercial Li-ion primary cell from Duracell using a combination of two highly complementary tomography methods. Tomography is a technique for displaying a representation of a cross section through a solid object through the use of a penetrating wave such as ultrasound or X-rays. The method is used in radiology, archaeology, atmospheric science, geophysics, oceanography as well as materials science.

X-rays are sensitive to heavier elements in the battery—such as manganese and nickel, and neutrons are sensitive to lighter elements—lithium and hydrogen, allowing the two techniques to visualize different parts of the battery structure and allowing researchers to build up a more complete understanding of the processes occurring deep within the cell during battery discharge.

X-ray computed tomography allowed for the quantification of mechanical degradation effects such as electrode cracking from the electrode bending process during cell manufacturing. Whereas the imaging using neutrons yielded information about the electrochemistry such as lithium-ion transport and consumption or gas formation by electrolyte decay.

A new mathematical method developed at the Zuse-Institut in Berlin then enabled researchers to virtually unwind the battery electrodes that are wound into the form of a compact cylinder. The cylindrical windings of the battery are difficult to examine quantitatively, and the cell cannot be unwound without inducing further damage that would not be present in an unwound battery.

The combination of X-ray and neutron imaging modalities provides clear benefits for the interpretation of active electrochemical materials, which contain many components with different X-ray and neutron attenuation properties. We expect that this methodology will be valuable for studying and optimizing a range of battery chemistries, not limited to the primary cell studied here; it will also make great benefits in the field of rechargeable cells.

Moreover, the unrolling protocols demonstrated here provide a new means for direct correlation and comparison of multi-modal image data, which aid the straightforward interpretation of spatial trends that are otherwise challenging in a spiral-wound configuration. Furthermore, these data present an example of the powerful insight that may be gained from 4D imaging; the bulk properties that computational models were previously based on may now be replaced by spatially resolved, transient values. The implications of this lie in that models would not need to apply the macroscopic assumptions made by averaging bulk properties, and consequently, due to the capability of incorporating local heterogeneities, they would be able to provide a more accurate and comprehensive description of the practical operation, degradation and failure of a cell.

—Ziesche et al.

Resources

• Ralf F. Ziesche, Tobias Arlt, Donal P. Finegan, Thomas M.M. Heenan, Alessandro Tengattini, Daniel Baum, Nikolay Kardjilov, Henning Markötter, Ingo Manke, Winfried Kockelmann, Dan J.L. Brett, & Paul R. Shearing (2020) “4D imaging of lithium-batteries using correlative neutron and X-ray tomography with a virtual unrolling technique.” Nat Commun 11, 777 doi: 10.1038/s41467-019-13943-3