KAIST researchers have identified a possible route toward electric-vehicle batteries that charge faster without losing as much performance or lifespan. Their approach uses a three-dimensional “digital twin” of a real graphite battery anode, allowing them to observe how microscopic variations inside an electrode can trigger lithium plating, uneven protective-film growth, mechanical stress, and eventual degradation.
The study, led by Professor Kang Taek Lee of the Department of Mechanical Engineering at the Korea Advanced Institute of Science and Technology, or KAIST, was conducted with Professor EunAe Cho of the Department of Materials Science and Engineering. Instead of treating a battery electrode as a uniform block with average properties, the researchers recreated its internal architecture in three dimensions, including graphite particles, binder material, and electrolyte-filled pores. The resulting model was designed to behave like a virtual counterpart of a commercial graphite anode.
That internal structure matters because a lithium-ion battery is not simply a container in which ions move smoothly from one side to another. During charging, lithium ions travel through the electrolyte-filled pores of the anode and enter graphite particles, where they are stored between layers of carbon atoms. When charging is too rapid, however, the ions may reach the graphite surface faster than they can be absorbed. Instead of intercalating into the graphite, they can accumulate as metallic lithium on the surface, a damaging process known as lithium plating.
Lithium plating is one of the most important obstacles to extreme fast charging. It can consume active lithium, reduce the battery’s usable capacity, and in some cases create structures that increase the risk of internal short circuits. At the same time, a thin protective layer called the solid electrolyte interphase, or SEI, forms on the graphite surface. The SEI is essential because it helps stabilize the electrode, but excessive or uneven growth consumes electrolyte and lithium, raises resistance, and can prevent ions from reaching the graphite efficiently.
The anode also undergoes mechanical changes during charging. As lithium enters graphite, the particles expand and push against neighboring particles, binder regions, and pore walls. If the surrounding structure has sufficient empty space, that expansion can be accommodated with less damage. If the local pore volume is too limited, mechanical stress becomes concentrated in particular regions. Because lithium transport, SEI growth, lithium plating, and mechanical deformation occur simultaneously at microscopic scales, experiments that measure only total capacity can miss the earliest signs of failure.
To expose these hidden processes, the KAIST team reconstructed the three-dimensional arrangement of the graphite particles, polymer binder, and pores in a commercial anode. The researchers then altered key structural parameters in the virtual electrode, including its thickness, porosity, and the spatial distribution of the binder. They simulated fast-charging conditions and tracked where lithium ions moved, where lithium plating occurred, how the SEI developed, and which areas experienced the greatest mechanical stress.
The simulations revealed that two electrodes with nearly identical overall compositions and apparent charging capabilities can behave very differently internally. In 50-micrometer-thick anodes, changing the binder distribution produced a capacity difference of less than 4 percent, a variation that might appear relatively minor in conventional battery testing. Yet the simulations showed clear differences in the locations where lithium was inserted into graphite and where degradation reactions were concentrated.
One particularly important result emerged when binder was concentrated near the separator, the membrane that separates the anode from the cathode while allowing lithium ions to pass. The binder occupied space that could otherwise support ion transport, effectively narrowing the pathways through which ions moved into the electrode. This created a microscopic bottleneck similar to traffic congestion on a narrowed road. Under those conditions, lithium plating near the current collector increased by more than 10 percent compared with an anode in which the binder was distributed more evenly.
A more uniform binder arrangement produced more consistent ion transport and helped the SEI form more evenly across the electrode. The contrast became substantially stronger as the anode grew thicker. In an 83-micrometer electrode, the difference in charge capacity between the two binder distributions reached approximately 18 percent. The finding highlights a growing challenge in battery engineering: thicker electrodes can store more energy per unit of area, but their greater transport distance makes them more sensitive to local variations in pores, binder, and particle arrangement.
The researchers also found that pore-space distribution influenced mechanical damage. Regions with adequate pore volume could absorb some of the expansion of graphite particles during charging, while densely packed areas forced particles against one another and developed concentrated stress. These localized effects may not immediately appear in a battery’s total voltage or capacity, but they can gradually accelerate structural damage and amplify other degradation mechanisms. The study therefore suggests that electrode design must consider not only how much graphite, binder, and pore space are present, but also their precise locations.
The digital-twin strategy could allow battery developers to test virtual electrode designs before producing and cycling large numbers of physical prototypes. By revealing where transport bottlenecks, lithium plating, uneven SEI growth, and mechanical stress are likely to occur, the model may help engineers optimize electrodes for fast charging while preserving energy density and service life. Professor Lee said the work demonstrates how three-dimensional modeling can uncover internal battery problems that remain invisible when researchers rely only on overall charging performance. The study, led by KAIST PhD candidate Yejin Kang as first author, was published in InfoMat and featured on the journal’s back cover. Its results point toward a future in which the microscopic architecture of an electrode is designed as carefully as its chemical ingredients.
Subject of Research: Three-dimensional digital-twin modeling of graphite lithium-ion battery anodes, fast-charging degradation, lithium plating, SEI formation, ion transport, binder distribution, pore structure, and mechanical stress.
Article Title: Digital twin quantifies spatial-heterogeneity-driven failure in fast-charging lithium-ion battery anodes
News Publication Date: August 24, 2026
Web References: https://doi.org/10.1002/inf2.70141
References: Kang, Y. et al., “Digital twin quantifies spatial-heterogeneity-driven failure in fast-charging lithium-ion battery anodes,” InfoMat, DOI: 10.1002/inf2.70141. Article publication date: July 7, 2026.
Image Credits: KAIST
Keywords
Lithium-ion batteries, electric vehicles, fast charging, battery degradation, lithium plating, graphite anodes, digital twins, solid electrolyte interphase, electrode microstructure, binder distribution, pore structure, battery materials, energy storage, KAIST.
