Lithium-ion battery chemistry and charge level shape how cells fail

Tests of commercial cells found that heat, gas release and fire spread varied with cathode type, cell design and charge level.

Editorial illustration — not from the study.

Researchers carried out more than 170 abuse tests on commercial lithium-ion cells with four cathode chemistries. They measured when thermal runaway began, peak cell temperature, gas volume and maximum overpressure in both single-cell tests and tests of failure spreading between cells.

Cells using nickel manganese cobalt oxide, lithium cobalt oxide or lithium nickel cobalt aluminum oxide showed different behavior from lithium iron phosphate cells. In the first group, gas release was linked to the energy stored at the tested charge level, while propagation usually occurred at 70–100% charge but not at 30%.

How the cells failed

The study tested 10 commercial lithium-ion cell types at several states of charge, in single-cell and propagation experiments. The researchers identified two broad groups based on cathode chemistry: cells using lithium manganese cobalt oxide, lithium cobalt oxide or lithium nickel cobalt aluminum oxide, and cells using lithium iron phosphate.

For the first group, emitted gas volume increased linearly with charge-dependent cell energy, at 0.46 liters per watt-hour, regardless of the surrounding test atmosphere. In propagation tests, the first group generally spread thermal runaway to neighboring cells at 70–100% state of charge, but typically did not spread it at 30%. Propagation was classified as non-, slow- or fast-propagating. The tests also compared onset temperature, maximum cell temperature and maximum overpressure.

Why charge and chemistry matter

The results suggest that some thermal-runaway effects may be estimated from accessible cell features such as format, cathode chemistry, capacity or energy and charge level. That could help researchers and engineers compare battery designs and develop safety measures based on how particular cells release heat and gas or spread failure. The findings also show that charge level is important when assessing propagation risk for the tested cell groups.

Tests and limits

This is an experimental study based on more than 170 abuse tests of 10 commercial cell types, covering single-cell and propagation scenarios. It directly measured thermal-runaway temperatures, gas release, overpressure and propagation behavior.

The abstract does not establish that the reported relationships apply to every lithium-ion cell, operating condition or accident scenario. The results are based on the tested commercial formats, chemistries, charge levels and experimental conditions, and the abstract does not provide enough detail to determine how accurately these features can predict outcomes for cells outside that set.

A simulated power grid predicted injected faults in under a second

Researchers developed a system that analyzes live data from grid sensors and other devices to predict faults in a simulated power network. In tests using a 33-bus IEEE network with injected faults, it achieved up to 96.4% accuracy while making predictions in less than a second.

5 minutes ago

Engineering & Technology

A smartphone strip reads tear glucose and pH in 30 seconds

Researchers developed a Schirmer strip that simultaneously indicates glucose and pH through color changes, with smartphone image analysis helping assess the results. In tests using artificial tear fluid, the glucose signal appeared within 30 seconds and had a visual detection limit of 0.5 mM.

15 hours ago·5 views

Engineering & Technology

A hydrogen process turns rare-earth magnet waste into reusable materials

Researchers have used hydrogen plasma to process Nd-Fe-B magnet machining waste, separating it into metallic iron and a rare-earth oxide-rich fraction. The process also formed rare-earth hydrides, creating feedstocks that could be used in further magnet production without acid.

1 day ago·12 views