White paper
Physics-based lifetime prediction
Calibrating degradation mechanisms to extrapolate cell aging
A physics-based degradation model calibrated on the first seven months of a public 13-month aging study of LG M50 cells predicts the remaining six with a mean capacity error of 3.2 percentage points, at 10, 25 and 40 °C. We compare it with an empirical fade law on the same data, show what the calibrated mechanisms say about why the cells aged and where the data stops telling them apart, and explain what it takes to make this calibration routine.
What you'll learn
- Why degradation modes, not capacity, are the right calibration target
- How a physics-based model and an empirical curve compare on the same held-out data
- What the calibrated mechanisms say about why these cells aged, and where the data runs out
- What makes calibrating these models expensive, and what it takes to make it routine
Who it's for
- Battery engineers planning aging campaigns
- Modelers calibrating lifetime models
- R&D leads
Dr. Michael McEldrew, Electrolyte Scientist, Anthro Energy
Ionworks has been an exceptional partner, bringing deep expertise, responsiveness, and a genuinely collaborative approach to how we move from materials innovation to real-world cell performance.See how we work with materials teams