CID operation and pressure release

Quantitative Battery Venting: A Worked Example

The Venting and Physics of Thermal Runaway pages describe cell venting qualitatively: gas builds up inside the cell, a burst disc or crimped vent opens once a threshold pressure is reached, and hot gas and electrolyte aerosol are ejected. This post works through the same event quantitatively, how fast the gas actually leaves the cell and what it’s made of, using a documented case study rather than invented numbers.

Choked Flow at the Burst Disc

Once the pressure ratio across the vent opening exceeds the critical ratio for the gas involved (roughly 1.9:1 for a diatomic-dominated mixture), the flow through the opening becomes choked: gas velocity at the throat reaches the local speed of sound and further increases in upstream pressure no longer increase the flow velocity, only the density. This is the flow regime a bursting cell operates in for most of its venting duration, since internal pressures at rupture are typically several bar above ambient.

The choked mass flow rate through an orifice of effective area A* is:

m˙=CdA∗P0γRT0(2γ+1)γ+12(γ−1)\dot{m} = C_d A^{*} P_0 \sqrt{\frac{\gamma}{R T_0}} \left(\frac{2}{\gamma+1}\right)^{\frac{\gamma+1}{2(\gamma-1)}}

where Cd is the discharge coefficient of the vent geometry, P0 and T0 are the stagnation (internal cell) pressure and temperature, R is the specific gas constant of the vented mixture, and γ is its ratio of specific heats. Sandia National Laboratories and New Mexico Tech measured Cd and effective opening area directly for commercial 18650 vent caps by instrumenting burst discs with stagnation and static pressure taps and back-calculating the two unknowns from conservation of mass under this same compressible-isentropic relationship — their results showed measurable variation in both discharge coefficient and opening area between vent cap manufacturers, and since mass flow rate scales linearly with both terms, that manufacturing variation translates directly into uncertainty on the predicted venting rate (Mier et al., Journal of Fluids Engineering, 2019).

Worked Example: An 18650 NCA Cell

A CFD study of cell-to-cell propagation in an 18650 module worked this through for a documented failure case rather than a generic assumption, and it’s a useful reference point because every number in it traces back to a measured source rather than a guess (Grimmeisen & Schneider, 13th European LS-DYNA Conference, 2021):

  • Burst disc opening area: 8 mm² (within the 6.111–9.773 mm² range measured across commercial 18650 vent caps)
  • Total mass lost during venting: 20.5 g, over an assumed 5 s venting duration, giving a mass flow rate of ~4.1 g/s
  • Vented gas temperature: ~896°C
  • Resulting choked exit velocity at the burst disc: ~473 m/s

That exit velocity is what actually drives cell-to-cell propagation risk inside a module: it’s the term that determines how quickly hot gas reaches a neighbouring cell’s surface and how much convective heat it delivers on arrival, which is why pack-level venting and gas-guidance design (channel cross-section, vent stack routing, distance to the nearest neighbour) has as much influence on propagation as the cell’s own onset temperature.

Gas Speciation

The vented mixture isn’t a single gas, and its composition changes the effective R and γ used above, as well as the flammability and toxicity risk downstream. Composition depends strongly on cathode chemistry and state of charge at failure. Gas chromatography on LFP cells at 75% SOC found H₂ as the largest single component (37.1% by volume), followed by CO₂ (31.3%), CH₄ (11.9%) and C₂H₄ (10.1%) — those four species alone accounting for over 90% of the total vented volume (study on SOC-gradient thermal runaway in LFP cells). NMC and NCA chemistries generally vent a broader mix of CO, CO₂, H₂ and volatile organics, with CO and CO₂ tracing back to cathode oxidation and H₂ to anode-side reactions with the electrolyte. Hydrogen fluoride is also present across chemistries, formed from reactions between the electrolyte’s LiPF₆ salt and trace moisture, and is a meaningful toxicity concern even though it’s typically a small fraction of total vented volume.

Because H₂, CO and light hydrocarbons dominate by volume, the vented mixture is both lighter than air on average and well within flammability limits once it mixes with ambient oxygen — the practical reason vent stack design in a pack has to route gas away from ignition sources and other cells rather than simply venting to the nearest gap in the enclosure.

References

  1. Mier, F.A.; Hargather, M.J.; Ferguson, S. Experimental Quantification of Vent Mechanism Flow Parameters in 18650 Format Lithium Ion Batteries, Journal of Fluids Engineering, 141(6), 2019, Sandia National Laboratories / New Mexico Institute of Mining and Technology
  2. Grimmeisen, D.; Schneider, M.S. Numerical Simulation of Cell Venting within a Simplified 18650 Li-Ion Battery Pack, 13th European LS-DYNA Conference, 2021, Cascate GmbH (citing mass-loss and duration data from Golubkov et al., RSC Advances, 5, 2015)
  3. Study on the Multi-Stage Evolution of Thermal Runaway and the Flammability Threshold of Gas Generation in Lithium Iron Phosphate Batteries Based on SOC Gradient

See also: Venting, Physics of Thermal Runaway

Leave a Comment