A group of researchers from various research centers, including the US Department of Energy’s SLAC National Accelerator Laboratory, Stanford University, the Toyota Research Institute, and MIT, have used computer vision, a form of machine learning, to gain a better understanding of how rechargeable lithium-ion batteries work.

The scientists examined X-ray films of battery electrodes at the nanometer level and discovered previously hidden physical and chemical details. This breakthrough has the potential to improve the efficiency of lithium-ion batteries and also have broader applications in understanding complex systems, such as cell division in embryos.

The study focused on lithium iron phosphate (LFP) particles, which are commonly found in the positive electrodes of lithium-ion batteries. These particles are coated with a thin layer of carbon to enhance electrical conductivity.

In order to observe the internal processes of the batteries, the research team created transparent battery cells with two electrodes surrounded by an electrolyte solution containing freely moving lithium ions. Through this setup, they were able to track the movement of lithium ions during charge and discharge cycles. This process, known as intercalation, involves the ions entering and leaving the LFP particles.

Lithium iron phosphate (LFP) is highly significant in the battery industry due to its low cost, safety record, and use of abundant elements, making it particularly relevant in the electric vehicle market.

This study was carried out as a collaboration between researchers that began eight years ago when Professor Martin Bazant from MIT and William Chueh from Stanford combined their expertise in mathematical modeling and advanced X-ray microscopy to study battery particles. Later on, they incorporated machine learning tools to accelerate battery testing and identify optimal charging methods. The current study takes it a step further by leveraging computer vision to analyze nanometer-level X-ray films from 2016, allowing for a more comprehensive understanding of lithium insertion reactions within the LFP particles.

By pixelating the X-ray images, the researchers were able to capture the concentration of lithium ions at each point within the particle. This enabled them to create films that illustrate the flow of lithium ions into and out of the particles during charge and discharge.

By analyzing the X-ray images, the researchers found that the movement of lithium ions within the material closely aligned with computer simulations previously developed by Bazant. They used 180,000 pixels as measurements to train a computational model that accurately describes the out-of-equilibrium thermodynamics and reaction kinetics of the battery material.

Furthermore, the study revealed that variations in lithium ion absorption on the surface of the particle are correlated with the thickness of the carbon coating. This finding suggests that optimizing the thickness of the carbon layer could improve battery efficiency, representing a significant advancement in battery design.

The results of this study provide insights for optimizing lithium iron phosphate electrodes and demonstrate the potential of machine learning and advanced imaging techniques in unraveling the mysteries of materials and systems. This breakthrough not only paves the way for improvements in battery technology but also holds promise for studying pattern formation in other chemical and biological systems.

– MIT Press Release
– Stanford University Press Release
– “Nature” Journal