Li-ion safety

  • Highest heat production and physical response is not correlated
  • Safety assessment based purely on heat evolution and cathodechemistry can give false safeties.
  • Safety is best assessed by a combination of mechanical stress, amount of electrical energy (energydensity, size and charge level) and chemical composition (How much heat can be generated)
  • Measurements show that cathode and anode materials are most chemically unstable when the cell is at 100% SOC (p12)(
  • Heatup initializing temperature usually occurs from SEI-reactions at the anode. (Usually between 75-124)\(^\circ\)C.
  • Thermal runaway usually comes from cathodereactions and start between 140-240\(^\circ\)C.
  • Thermal response to degradation mechanisms (NMC+LMO 1:1 tandem 18650 cell, Fleischhammer et al.):
    • Calendar aging: minimal change
    • High C-rate cycling: 8% lower heatup initialization temp
    • Low temp (-10\(^\circ\)C): 31/54\(^\circ\)C init temp in stead of 90/105\(^\circ\)C + more heat evolution in total.

Source: M.Sc. degree of Torleif Lian (FFI)

ARC VS DSC

DSC: Differential scanning calorimetry.

  • Two small containers, one with sample, one empty
  • Heated in a chamber (5\(^\circ\)C/min), if edo or exothermic reactions happen in the sample, there will be a heat flux difference from the reference container.
  • Bad:
    • Not suitable for large samples, like a battery
    • Simultaneous pressure measurements not possible
    • less sensitive then ARC
    • Gives higher onset temps.

ARC:

  • bad:
    • affected by sample/container weight ratio, which can comprise sensitivity
logo