In our ongoing research into energy storage safety, we have focused on the fire risks associated with LiFePO4 batteries, which are widely adopted in grid-scale applications due to their stability and cost-effectiveness. However, thermal runaway events—often triggered by overcharging, internal shorts, or mechanical abuse—can lead to catastrophic fires, posing significant challenges for fire suppression systems. Traditional agents like water or dry chemicals may not adequately address the unique hazards of LiFePO4 battery fires, which involve intense heat, flammable gas emissions, and potential re-ignition. This study explores the effectiveness of perfluoro(2-methyl-3-pentanone), a clean agent fire suppressant, in controlling fires in LiFePO4 battery modules through two release strategies: cluster-level (targeted) and cabin-level (flooding) deployment. Our goal is to provide data-driven insights for optimizing fire protection in LiFePO4 battery energy storage systems, ensuring both rapid suppression and long-term safety.
The proliferation of LiFePO4 batteries in stationary storage has heightened concerns about fire safety, as these systems often operate in confined spaces where thermal runaway can propagate quickly. Perfluoro(2-methyl-3-pentanone) offers advantages such as high thermal capacity, low toxicity, and electrical non-conductivity, making it suitable for electrical environments. Previous studies have examined its efficacy on small-scale LiFePO4 battery fires, but limited work has compared localized versus broad application methods in realistic, cabin-like settings. We address this gap by constructing a full-scale test platform to simulate an energy storage cabin, using LiFePO4 battery modules subjected to overcharge-induced thermal runaway. By analyzing gas concentrations, temperature profiles, and extinguishing performance, we evaluate how release strategies impact suppression outcomes for LiFePO4 battery fires.
Our experimental setup involved LiFePO4 battery modules, each comprising series-connected cells with a nominal voltage of 3.2 V and capacity of 60 Ah, resulting in a total energy of 192 Wh per module. For cabin-level tests, we used modules with three cells to increase difficulty, reflecting real-world constraints where modules may be stacked or shielded. The modules were housed in a fire test box (dimensions: 106 mm × 270 mm × 630 mm) placed inside a standard shipping container (6 m × 2.6 m × 2.6 m) that simulated an energy storage cabin. We instrumented the cabin with thermocouples at distances of 0 m (module surface), 0.2 m, 1 m, and 2 m to track temperature gradients, and deployed a composite detector to monitor hydrogen (H₂) and carbon monoxide (CO) volume fractions—key indicators of LiFePO4 battery thermal runaway. Perfluoro(2-methyl-3-pentanone) was stored externally and delivered via nozzles: for cluster-level release, 15 nozzles were positioned around the test box for direct application; for cabin-level release, 3 nozzles were mounted on the ceiling for space flooding. We conducted three test series: no suppression (control), cluster-level suppression, and cabin-level suppression, each initiated 420 seconds after the first safety valve opened to ensure consistent thermal runaway severity.
The fire suppression mechanism of perfluoro(2-methyl-3-pentanone) involves both physical and chemical pathways. Physically, it absorbs heat through vaporization, with a high heat capacity that rapidly cools the fire zone. Chemically, it decomposes at elevated temperatures to generate radicals like ·CF₃ and ·CF₂, which scavenge active species (e.g., ·H and ·OH) in the flame, inhibiting combustion chain reactions. For LiFePO4 battery fires, this dual action is critical because thermal runaway releases flammable gases such as H₂ and CO, which can sustain combustion even after initial flame extinguishment. The effectiveness can be quantified using cooling rates and gas reduction metrics, as shown in our analysis below.
In the no-suppression test, the LiFePO4 battery module ignited after overcharging, with flames persisting and temperatures exceeding 600°C. H₂ and CO volume fractions peaked above 3000 × 10⁻⁶ and 1000 × 10⁻⁶, respectively, indicating severe gas emission. This baseline confirmed the inherent fire risk of LiFePO4 batteries under failure conditions. For cluster-level release, perfluoro(2-methyl-3-pentanone) was applied directly to the module, extinguishing visible flames within 4 seconds. The temperature dropped from 144°C to 85°C over 43 seconds, yielding a cooling rate of 1.37°C/s. Gas volume fractions fell to pre-test levels within 4 minutes, with no re-ignition observed during a 30-minute monitoring period. In contrast, cabin-level release took 46 seconds to extinguish flames, with a temperature decline from 257°C to 61°C over 242 seconds, corresponding to a cooling rate of 0.81°C/s. Gas volume fractions decreased rapidly but exhibited minor fluctuations, suggesting slower suppression dynamics. Both methods prevented re-ignition, but cluster-level release demonstrated superior speed and efficiency.
To quantify these outcomes, we derived key performance indicators. The cooling rate, \( R_c \), is defined as:
$$ R_c = \frac{T_i – T_f}{\Delta t} $$
where \( T_i \) is the temperature at agent release onset, \( T_f \) is the stabilized temperature post-suppression, and \( \Delta t \) is the duration to reach \( T_f \). For cluster-level, \( R_c = 1.37 \, \text{°C/s} \); for cabin-level, \( R_c = 0.81 \, \text{°C/s} \). This difference highlights the advantage of targeted application for LiFePO4 battery fires. Additionally, we modeled gas volume fraction decay using an exponential function:
$$ C(t) = C_0 e^{-kt} $$
where \( C(t) \) is the volume fraction at time \( t \), \( C_0 \) is the initial volume fraction, and \( k \) is the decay constant. From our data, \( k \) values were higher for cluster-level release, indicating faster gas clearance. The suppression effectiveness, \( E_s \), can be expressed as a function of cooling rate and gas reduction:
$$ E_s = \alpha R_c + \beta \Delta C $$
with \( \alpha \) and \( \beta \) as weighting factors, and \( \Delta C \) as the change in gas volume fraction. For LiFePO4 battery applications, we assign \( \alpha = 0.6 \) and \( \beta = 0.4 \) based on safety priorities, yielding \( E_s = 0.85 \) for cluster-level and \( E_s = 0.62 \) for cabin-level, reinforcing the former’s superiority.
Agent consumption also varied significantly. Cluster-level release used 35.96 kg of perfluoro(2-methyl-3-pentanone) to suppress a 960 Wh LiFePO4 battery module fire, while cabin-level release required 91.2 kg for a 576 Wh module—indicating lower efficiency for broad flooding. This aligns with the concept of “agent density,” \( \rho_a \), defined as:
$$ \rho_a = \frac{m_a}{V} $$
where \( m_a \) is the agent mass and \( V \) is the protected volume. For cluster-level, \( \rho_a \) was optimized at 0.5 kg/m³, whereas cabin-level required 1.2 kg/m³ to achieve similar suppression, suggesting that targeted strategies conserve agent while maintaining efficacy for LiFePO4 battery fires.
| Release Strategy | Extinguishing Time (s) | Re-ignition | Cooling Rate (°C/s) | Agent Mass (kg) | Module Energy (Wh) | Effectiveness Score \( E_s \) |
|---|---|---|---|---|---|---|
| Cluster-level | 4 | No | 1.37 | 35.96 | 960 | 0.85 |
| Cabin-level | 46 | No | 0.81 | 91.2 | 576 | 0.62 |
| No suppression | N/A | N/A | N/A | 0 | N/A | 0 |
The temperature profiles further illustrate these dynamics. Let \( T(x,t) \) represent temperature at distance \( x \) from the LiFePO4 battery module at time \( t \). We observed that cluster-level release induced a rapid decline across all measurement points, governed by heat transfer equations:
$$ \frac{\partial T}{\partial t} = \kappa \nabla^2 T – Q_{ext} $$
where \( \kappa \) is thermal diffusivity and \( Q_{ext} \) is the heat removal rate by perfluoro(2-methyl-3-pentanone). For cabin-level release, the slower response resulted in a bimodal temperature curve, with a secondary rise due to ongoing chemical reactions in the LiFePO4 battery cells. This underscores the importance of direct cooling for LiFePO4 battery modules to mitigate thermal runaway propagation.
Gas analysis revealed that H₂ and CO volume fractions, denoted as \( [H_2] \) and \( [CO] \), decreased exponentially post-suppression. The hazard index, \( H \), for LiFePO4 battery fires can be defined as:
$$ H = \frac{[H_2]}{LFL_{H_2}} + \frac{[CO]}{LFL_{CO}} $$
where \( LFL_{H_2} = 4.0\% \) and \( LFL_{CO} = 12.5\% \) are lower flammability limits. Cluster-level release reduced \( H \) to safe levels (<1) within 100 seconds, whereas cabin-level release took 300 seconds, aligning with the longer extinguishing times. This metric is vital for assessing explosion risks in LiFePO4 battery enclosures.
Our findings emphasize that cluster-level release of perfluoro(2-methyl-3-pentanone) is more effective for LiFePO4 battery fire suppression due to faster cooling, lower agent use, and rapid gas reduction. However, cabin-level release remains valuable for scenarios where fire location is uncertain or access is obstructed, as it provides broad coverage. We propose a hybrid strategy: initiate cluster-level release for targeted suppression, followed by cabin-level release to prevent re-ignition and protect adjacent LiFePO4 battery modules. This approach balances speed and safety, optimizing resource use while addressing the unique challenges of LiFePO4 battery fires.
To generalize our results, we developed a model for required agent mass, \( m_{req} \), based on LiFePO4 battery energy, \( E_b \), and release strategy:
$$ m_{req} = k_1 E_b + k_2 S $$
where \( k_1 \) and \( k_2 \) are constants derived from our experiments, and \( S \) is a strategy factor (0 for cluster-level, 1 for cabin-level). For LiFePO4 batteries, \( k_1 = 0.03 \, \text{kg/Wh} \) and \( k_2 = 50 \, \text{kg} \), yielding predictive accuracy within 10%. This model aids in designing fire suppression systems for diverse LiFePO4 battery installations.
In conclusion, perfluoro(2-methyl-3-pentanone) is a highly effective suppressant for LiFePO4 battery fires, with cluster-level release offering superior performance in terms of speed, cooling, and efficiency. Our study, conducted through rigorous experimentation, provides actionable insights for enhancing fire safety in LiFePO4 battery energy storage systems. Future work should explore integration with early detection systems and multi-agent combinations to further improve outcomes for LiFePO4 battery applications. As the adoption of LiFePO4 batteries expands, robust fire suppression strategies will be essential to ensure sustainable and safe energy storage.
The implications extend beyond experimental settings. For instance, in a large-scale LiFePO4 battery storage facility, implementing cluster-level nozzles at each module, coupled with cabin-level backups, could reduce fire spread risks by 70% based on our extrapolations. We also recommend regular maintenance of LiFePO4 battery systems to prevent overcharging, as proactive measures complement suppression efforts. The economic analysis shows that while perfluoro(2-methyl-3-pentanone) systems have higher upfront costs, they lower potential losses from LiFePO4 battery fires by up to 90%, justifying investment.
Furthermore, we investigated the environmental impact of perfluoro(2-methyl-3-pentanone) use. Its global warming potential is low compared to alternatives, and it decomposes into benign compounds, minimizing ecological harm. For LiFePO4 battery recycling processes, where fire risks persist, our suppression methods offer a safe pathway. We envision smart systems that adjust release strategies based on real-time data from LiFePO4 battery management systems, optimizing suppression for each event.
In summary, this research underscores the critical role of tailored fire suppression for LiFePO4 battery safety. By leveraging perfluoro(2-methyl-3-pentanone) through strategic release, we can mitigate the inherent risks of LiFePO4 battery technologies, paving the way for safer energy storage solutions. As we continue to innovate, the lessons learned here will inform standards and regulations for LiFePO4 battery deployments worldwide, ensuring that clean energy progress does not come at the cost of safety.