Badabinski

You’re correct that they can enter thermal runaway, they just can’t autoignite. I really suspect that if this site has been using LiFePO4 cells instead of NMC, it wouldn’t have gone up like it did. 3000 MWh of NMC cells sounds absolutely bugnuts crazy to me.

This is why you don’t use battery chemistries that can thermally run away autoignite in grid storage. The plant was using LG JH4 batteries, which use an NMC chemistry. I don’t think that LiFePO4 cells were as ubiquitous when this plant was first constructed, so the designers opted for something spicy instead.

This shit is why you use LiFePO4. It can’t thermally run away autoignite, it lasts longer, and the reduced energy density doesn’t really matter for grid storage. Plus, it doesn’t use nickel or cobalt so the only conflict resource is lithium.

EDIT: LiFePO4 batteries can enter thermal runaway, but they can’t autoignite.

We lack the materials and engineering necessary to make lifted weight storage systems enter the order of magnitude of energy storage needed to compete with batteries, let alone pumped hydro. It’s just really, really hard to compete with literal megatons of water pumped up a 500 meter slope.

I believe that the plant in question was using something besides Lithium Iron Phosphate batteries. This press release mentions LG JH4 which are deffo not LiFePO4. LiFePO4 batteries are far, far safer than other Lithium chemistries, and are now the norm for BESS (not cars tho, since they have lower energy density but better a better lifetime than NMC/NCA). This fire would not have happened with a BESS using LiFePO4 batteries.

Now that batteries with aqueous sodium-ion chemistries are becoming available, we should begin transitioning pre-LiFePO4 sites to those wholesale. Aqueous sodium-ion batteries should be even safer than LiFePO4, and while they have kinda shit energy density, they’re still fine for grid storage.

EDIT: correction, LiFePO4 batteries can run away, but they are incapable of autoignition.