Utilities and developers around the world are deploying battery energy storage systems (BESS) at a massive scale to stabilize the grid.
For fire chiefs, risk assessors and facility managers, this rapid expansion represents a shifting threat landscape.
99% of these systems rely on lithium-ion chemistry- a technology that offers high energy density but introduces significant hazards when things go wrong.
To maintain safety, lithium-ion systems require a strictly controlled environment reliant on a complex interplay of mechanical components.
But, over a 20-year lifespan, no industrial system works perfectly 100% of the time. Cooling systems leak, sensors drift and parts degrade.
Graeme Grant
For most industrial products, a mechanical failure just means the machine stops working.
For lithium-ion batteries, a routine failure for the battery cell itself or one of its system components can trigger thermal runaway- a self-perpetuating fire that standard suppression systems struggle to contain.
The critical question for the fire safety industry isn’t how many sensors to install or how many protective barriers to mandate.
It is more fundamentally whether the battery industry can ever engineer a truly safe system around a chemistry that can combust when failures happen.
And they will happen.
Battery manufacturers often point out that lithium-ion cells themselves rarely fail unprovoked.
They’re right – the failure rate of lithium-ion batteries from an inherent defect is < 0.1 ppm (1 cell in 10 to 40 million).
However, data shows that most major BESS fires start with malfunctions of “balance of plant” (BOP) components; the inverters, coolant pumps, battery management systems, even fire suppression systems.
Graeme Grant
These are routine equipment failures that happen in industrial facilities every day.
What makes lithium-ion different is how catastrophically these routine failures escalate.
A faulty sensor or cooling leak shouldn’t trigger multi-day hazmat responses and neighborhood evacuations.
Blaming the incident on a cooling leak misses the point entirely, it’s like saying a grenade explosion wasn’t caused by the grenade, but by the pin falling out.
The chemistry itself becomes the amplifier.
During thermal runaway, the cathode releases oxygen that fuels a self-sustaining reaction with the flammable electrolyte.
Water can cool the battery’s exterior but cannot stop the internal chemical reaction, requiring massive volumes and sustained, multi-day application to prevent spread.
A 2024 analysis on BESS failures by the Electric Power Research Institute (EPRI) found that 89% of battery fires were caused by something other than the battery.
Graeme Grant
This trend makes it clear that mechanical or software failures can escalate into prolonged emergency events.
A few examples include:
Warwick, NY (June 2023): Moisture infiltrated a Lithium Iron Phosphate (LFP) facility near a school during a rainstorm, causing electrical shorts. The site burned for a week, forcing school closures
Moss Landing, CA (Sep. 2022): Rainwater entered through a dislodged umbrella valve at the Elkhorn battery facility, causing electrical arcing. Thermal runaway occurred in one BESS unit, requiring emergency response and temporary facility shutdown
Victoria, Australia (July 2021): A cooling system leak during commissioning caused a short circuit in an electronic component. The resulting thermal runaway spread to an adjacent compartment, consuming two Tesla Megapacks
In each case, the trigger was a foreseeable equipment malfunction. What turned these maintenance issues into week-long hazmat operations was the flammability hazard present underneath.
It’s worth noting that for more recent fire events, root cause analysis (RCA) is still ongoing due to insufficient information, in part due to damage caused during the event.
The industry is currently transitioning from high-nickel lithium-ion to iron-based LFP because it has a higher temperature threshold for thermal runaway.
As the Warwick incident proved, LFP still burns once compromised.
Furthermore, LFP fires generate significant amounts of hydrogen and toxic off-gases, shifting the hazard from purely thermal to a combined toxicity and explosion risk.
Graeme Grant
For firefighters, the operational challenge remains the same- inability to suppress directly and extended on-scene time, even if the nature of the hazards has shifted.
To truly mitigate risk, we need to look beyond response strategies and toward intrinsic safety.
Sodium-ion leverages a similar working principle as lithium-ion but offers a categorical departure from the risks:
Higher thermal runaway threshold: Thermal runaway in sodium-ion cells occurs at higher temperatures and progresses more slowly than in lithium-ion. Lithium-ion batteries can enter thermal runaway at temperatures as low as 110°C and spike to 900°C, whereas sodium-ion batteries typically require 160-180°C and peak at 300°C
Wider temperature range: Sodium-ion’s -40°C to 60°C operating range (versus 0°C to 45°C for lithium-ion) reduces reliance on active thermal management. Fewer components means lower balance-of-plant costs, reduced complexity and fewer failure points
Safe transport: Lithium-ion batteries must be shipped at 30% charge to avoid chemical and structural damage. This poses hazards during each phase of the delivery and install process. Sodium-ion batteries can be discharged to 0% state of charge, allowing them to be transported as electrically inert equipment
The safety argument for sodium-ion isn’t that system-level components won’t fail – they inevitably will. The argument is that when they do fail, the battery undergoes a completely different failure pathway
The newest generation of sodium-ion batteries (like the ones we’ve developed at Alsym) are designed to eliminate the risk of entering thermal runaway altogether.
Graeme Grant
Without that specific failure mechanism present, battery storage systems in the field become infinitely safer than the status quo.
As BESS facilities grow ubiquitous and move closer to population centers, our metrics for success must evolve to include system failure resilience.
The data from Moss Landing and Victoria confirms that while cells rarely fail, system components eventually will.
The difference between whether that failure becomes routine maintenance or a multi-agency emergency depends on the flammability profile of the battery.
Graeme Grant
The same EPRI analysis states BESS failure rates dropped 97% between 2018 to 2023 from approximately 9.2 failures per GW deployed to 0.22 per GW.
While this is certainly an improvement, the sheer scale of deployment growth means we will see more lithium-ion fires in absolute terms.
If the globe achieves the 2 TW of battery storage capacity BloombergNEF forecasts by 2035, 0.22 per GW still represents hundreds of battery failures and fires.
For fire protection professionals, promoting chemistries that don’t combust when the supporting systems fail is as critical as any mitigation strategy.
With sodium-ion, we can change the common denominator and move toward a standard where a coolant leak is just a leak, not a disaster.
