Skip to main content

Degradation of Li-ion batteries depends on how many times you’ve charged them

28-04-2022

How quickly a battery electrode decays depends on properties of individual particles in the battery – at first. Later on, the interaction of particles matters more. Scientists led by SLAC, Purdue and Virginia Tech (US) have just published these results in Science.

  • Share

Rechargeable lithium-ion batteries don't last forever – after enough cycles of charging and recharging, they'll eventually die, so researchers are constantly looking for ways to squeeze a little more life out of their battery designs.

Now, researchers at the SLAC National Accelerator Laboratory (US) and colleagues from Purdue University (US), Virginia Tech (US), and the ESRF have discovered that the factors behind battery decay actually change over time. Early on, decay seems to be driven by the properties of individual electrode particles, but after several dozen charging cycles, it's how those particles are put together that matters more.

"The fundamental building blocks are these particles that make up the battery electrode, but when you zoom out, these particles interact with each other," says SLAC scientist Yijin Liu, a researcher at the Stanford Synchrotron Radiation Lightsource and the corresponding author of the publication. "Therefore, if you want to build a better battery, you need to look at how to put the particles together”, he adds.

Seeing the forest for the trees

The new study builds on past research in which Liu and colleagues used computer vision techniques to study how the individual particles that make up a rechargeable battery electrode break apart over time. The goal this time was to study not just individual particles but the ways they work together to prolong – or degrade – battery life.

Keije Zhao, a Purdue mechanical engineering professor and co-author of the work, likened the problem to people working in groups. "Battery particles are like people – we all start out going our own way," Zhao says "But eventually we encounter other people, and we end up in groups, going in the same direction. To understand peak efficiency, we need to study both the individual behavior of particles, and how those particles behave in groups."

 With this aim, the team used X-ray nanotomography at the ESRF to reconstruct three-dimensional pictures of the cathodes after they had gone through either 10 or 50 charging cycles. They cut up those 3D pictures into a series of 2D slices and used computer vision methods to identify particles.

In the end, they identified more than 2,000 individual particles, for which they calculated not only individual particle features such as size, shape and surface roughness but also more global traits, such as how often particles came into direct contact with each other and how varied the particles' shapes were.

Next, they looked at how each of those properties contributed to particles’ breakdown, and a striking pattern emerged. After 10 charging cycles, the biggest factors were individual particles’ properties, including how spherical the particles were and the ratio of particle volume to surface area. After 50 cycles, however, pair and group attributes – such as how far apart two particles were, how varied their shapes were and whether more elongated, football-shaped particles were oriented similarly – drove particle breakdown. Peter Cloetens, scientist in charge at the ESRF’s ID16A, says: “The contribution of machine learning approaches has been key in revealing patterns hidden in the huge amount of data acquired at the ESRF.”

New clues for manufacturers

"It's no longer just the particle itself. It's particle-particle interactions that matter", Liu states. That's important, he said, because it means manufacturers could develop techniques to control such properties. For example, they might be able to use magnetic or electric fields to align elongated particles with each other, which the new results suggest would result in longer battery life.

The results could be applied beyond the particulars of the present research, as corresponding author and Virginia Tech chemist Feng Lin explains: "This study really sheds light on how we can design and manufacture battery electrodes to obtain long cycle life for batteries. We are excited to implement the understanding to next-generation, low-cost, fast charging batteries."

Reference:

Jizhou Li et al., Science, 28 April 2022, DOI: 10.1126/science.abm8962

Text by Nathan Collins (edited). 

 

Top image: The figure shows a piece of battery cathode after 10 cycles. The ML-based feature identification and quantification algorithms allow us to automatically single out the particles of interest, which are the severely damaged ones highlighted in the image. Credits: Yijin Liu