Peter Attia1 Aditya Grover2 Norman Jin1 Jerry Liao1 Peter Weddle4 Kristen Severson3 Michael Chen1 Nicholas Perkins1 Patrick Herring5 Muratahan Aykol5 Stephen Harris6 Robert Kee4 Richard Braatz3 Stefano Ermon2 William Chueh1

1, Materials Science & Engineering, Stanford University, Stanford, California, United States
2, Computer Science, Stanford University, Stanford, California, United States
4, Mechanical Engineering, Colorado School of Mines, Golden, Colorado, United States
3, Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
5, Toyota Research Institute, Los Altos, California, United States
6, Lawrence Berkeley National Laboratory, Berkeley, California, United States

The development of extreme fast charging (XFC) protocols (<10 minute charge time) for lithium-ion batteries is critical for widespread adaptation of electric vehicles and other devices with time-sensitive applications. However, a limited understanding of degradation modes during XFC and the large manufacturing variability of commercial lithium-ion batteries are major challenges to the development of high-performing XFC protocols. In this work, we perform both “top down” optimization and “bottom up” characterization of extreme fast charging across multiple length scales. First, we employ Bayesian optimal experimental design to develop a six-step XFC protocol for commercial 18650 lithium-ion batteries, achieving 80% state of charge in ten minutes. Next, we rationalize their high performance via cell-level simulations and electrode-level physical characterization. Our results reveal that loss of active material from the graphitic negative electrode is a significant source of degradation, preceding more conventional fast charging degradation modes such as lithium plating. We also identify a critical charge rate beyond which the loss of active material, and thus the overall cell degradation, accelerates. Finally, we study the rate of SEI growth during fast charging via electrochemical methods, identifying a strong dependence on the applied C rate. This work provides novel insight into optimizing and characterizing extreme battery fast charging for time-sensitive applications and suggests avenues to improve both charging times and lifetimes during aggressive battery operation.