In the context of escalating demand for electrochemical energy storage, sodium-ion batteries have emerged as a promising alternative to lithium-ion batteries due to their abundant raw material resources, lower cost, and comparable performance metrics. However, as the deployment of sodium-ion battery systems expands, particularly in large-scale storage applications, safety concerns pertaining to thermal runaway events have become increasingly salient. Thermal runaway, characterized by an uncontrolled temperature rise resulting from exothermic chain reactions within the cell, poses significant risks of fire, explosion, and system failure. This work presents a comprehensive experimental study aimed at elucidating the thermal runaway behavior, gas emission characteristics, and associated hazards of commercial large-capacity sodium-ion batteries under overheating conditions. By employing controlled heating tests and constant-volume pressure vessel experiments, we systematically analyze two distinct sodium-ion battery chemistries: one utilizing a layered oxide cathode (NaNi1/3Fe1/3Mn1/3O2) and the other a phosphate-based cathode (Na3V2(PO4)3). The findings provide critical insights into the thermal stability, gas generation profiles, and explosion risks, thereby contributing to the foundational safety assessment necessary for the secure integration of sodium-ion battery technology in energy storage systems.
The inherent safety of any battery chemistry is paramount, and for sodium-ion batteries, understanding the triggers and consequences of thermal failure is essential. Unlike lithium, sodium offers geographical abundance and reduced supply chain constraints, yet the fundamental electrochemistry shares similarities that warrant careful scrutiny of failure modes. Thermal runaway in sodium-ion batteries can be initiated by various abuse conditions, including thermal, electrical, and mechanical stresses. This study focuses on thermal abuse, simulating scenarios such as external heating or internal short circuits, to quantify the response of large-format cells. We investigate key parameters including onset temperature, maximum surface temperature, gas composition, pressure evolution, and flammability limits. The comparative analysis between the two cathode materials reveals significant differences in hazard severity, which can inform material selection, battery design, and safety protocol development for sodium-ion battery packs.
The experimental methodology encompasses two primary setups: a thermal runaway characterization test under open atmosphere and a gas analysis test within a sealed constant-volume vessel. For the thermal runaway tests, sodium-ion battery samples at 100% state of charge (SOC) are subjected to a constant heating power of 500 W via a heating plate. Temperature measurements are taken at multiple points on the cell surface and in the gas plume, while voltage, mass loss, and expansion force are monitored in real-time. The thermal runaway onset is defined as the moment when the surface temperature rise rate exceeds 1 °C/s for a duration of 10 seconds. For gas analysis, the same sodium-ion battery types are placed inside a 140 L pressure vessel filled with an inert atmosphere (less than 1% air). Upon thermal runaway triggered by heating, the internal pressure is recorded, and gas samples are collected for compositional analysis via gas chromatography-mass spectrometry (GC-MS). The ideal gas law is applied to calculate the total gas yield, and the lower explosion limit (LEL) of the gas mixture is determined using the Le Chatelier’s law.
The sodium-ion battery samples used in this study are commercial large-capacity prismatic cells, with specifications detailed in Table 1. Both sodium-ion battery types exhibit similar physical dimensions and capacities, enabling a direct comparison of their thermal behavior based on cathode chemistry. The NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery has a nominal capacity of 220 Ah, while the Na3V2(PO4)3 sodium-ion battery is rated at 230 Ah. All cells are conditioned to 100% SOC prior to testing to ensure consistency in energy content and reactivity.
| Parameter | NaNi1/3Fe1/3Mn1/3O2 Sodium-Ion Battery | Na3V2(PO4)3 Sodium-Ion Battery |
|---|---|---|
| Nominal Capacity | 220 Ah | 230 Ah |
| Dimensions (L × W × H) | 173.0 mm × 72.0 mm × 204.0 mm | 174.2 mm × 72.0 mm × 204.2 mm |
| Mass | 4900 ± 500 g | 5100 ± 200 g |
| Nominal Voltage | 3.1 V | 2.9 V |
| Charge/Discharge Voltage Range | 2.0 – 3.85 V | 2.0 – 3.85 V |
| Cathode Material | NaNi1/3Fe1/3Mn1/3O2 | Na3V2(PO4)3 |
| Anode Material | Hard Carbon | Hard Carbon |
Upon initiating the heating process, the thermal runaway event for both sodium-ion battery types unfolds through four distinct phases: the heating phase, the venting phase, the thermal runaway phase, and the decay phase. During the heating phase, the cell temperature rises steadily due to external heat input. Internal reactions begin with the decomposition of the solid electrolyte interphase (SEI) layer on the anode, followed by reactions between exposed sodium species and the electrolyte, generating flammable gases. As pressure builds, the safety valve opens, marking the venting phase where white smoke and少量 electrolyte are ejected. The thermal runaway phase is characterized by a violent ejection of large volumes of white gas, accompanied by occasional sparks, though no sustained ignition is observed due to the high ejection velocity. Finally, in the decay phase, gas emission subsides, and the cell temperature gradually decreases, though the aluminum casing may melt and pool at the bottom due to temperatures exceeding 1000 °C.
The temperature and voltage profiles during thermal runaway provide critical insights into the thermal stability and reaction kinetics of the sodium-ion battery. Figure 4 (not shown here, but described) illustrates the temporal evolution for both sodium-ion battery types. The NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery exhibits a lower onset temperature (Tonset = 266.3 °C) and a shorter time to onset (tonset = 321 s) compared to the Na3V2(PO4)3 sodium-ion battery (Tonset = 284.4 °C, tonset = 398 s). This indicates that the layered oxide cathode sodium-ion battery is more susceptible to thermal abuse. Moreover, the maximum surface temperature (Tmax) reached by the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery is 1082.9 °C, significantly higher than the 996.0 °C for the phosphate-based sodium-ion battery, suggesting a more intense exothermic reaction. The internal thermal runaway propagation rate (vTRP) is a key metric for assessing how rapidly the failure propagates within the cell. It is calculated using the formula:
$$ v_{\text{TRP}} = \frac{\delta}{\Delta t_{\text{TRP}}} $$
where \(\delta\) is the cell thickness (in mm) and \(\Delta t_{\text{TRP}}\) is the internal propagation time (in seconds). For the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery, with \(\delta = 72.0\) mm and \(\Delta t_{\text{TRP}} = 88\) s, the propagation rate is:
$$ v_{\text{TRP}} = \frac{72.0}{88} \approx 0.818 \text{ mm/s} $$
For the Na3V2(PO4)3 sodium-ion battery, \(\delta = 72.0\) mm and \(\Delta t_{\text{TRP}} = 105\) s, yielding:
$$ v_{\text{TRP}} = \frac{72.0}{105} \approx 0.686 \text{ mm/s} $$
The faster propagation rate in the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery underscores its higher thermal hazard potential. A summary of thermal characteristics is presented in Table 2.
| Battery Type | Onset Temperature, Tonset (°C) | Time to Onset, tonset (s) | Maximum Temperature, Tmax (°C) | Internal Propagation Time, ΔtTRP (s) | Thermal Runaway Propagation Rate, vTRP (mm/s) |
|---|---|---|---|---|---|
| NaNi1/3Fe1/3Mn1/3O2 Sodium-Ion Battery | 266.3 | 321 | 1082.9 | 88 | 0.818 |
| Na3V2(PO4)3 Sodium-Ion Battery | 284.4 | 398 | 996.0 | 105 | 0.686 |
The mechanical responses, including mass loss and expansion force, further elucidate the severity of thermal runaway in sodium-ion batteries. Mass loss reflects the ejection of volatile components and solids, while expansion force indicates internal pressure buildup prior to venting. Figure 5 shows the mass loss and mass loss rate curves. The total mass loss for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery is 1240.2 g, whereas the Na3V2(PO4)3 sodium-ion battery loses 1406.3 g. The higher mass loss in the phosphate-based sodium-ion battery suggests greater material ejection, potentially due to more vigorous gas generation. The peak mass loss rate, which correlates with the jet force during venting, is 123.2 g/s for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery and 178.5 g/s for the Na3V2(PO4)3 sodium-ion battery. This can be linked to the pressure dynamics using a modified Bernoulli equation, where the gas ejection velocity \(v_{\text{gas}}\) is related to the pressure \(P_{\text{gas}}\) and density \(\rho_{\text{gas}}\):
$$ v_{\text{gas}} = \varepsilon \sqrt{\frac{2P_{\text{gas}}}{\rho_{\text{gas}}}} $$
Here, \(\varepsilon\) is a discharge coefficient. The instantaneous mass loss peak corresponds to high \(P_{\text{gas}}\) and thus high \(v_{\text{gas}}\), indicating a forceful venting event.
Expansion force measurements, depicted in Figure 6, reveal that the Na3V2(PO4)3 sodium-ion battery experiences a maximum expansion force of 9.90 kN, nearly 1.5 times that of the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery (6.77 kN). Notably, the expansion force profile for the Na3V2(PO4)3 sodium-ion battery exhibits two distinct peaks, corresponding to two intense gas generation events likely associated with sequential thermal runaway in different parts of the cell core. This dual-peak behavior underscores the complex internal reaction dynamics in this sodium-ion battery chemistry.
Gas emission behavior is a critical aspect of sodium-ion battery safety, as the released gases can be flammable and explosive. The gas temperature in the plume, measured at various heights, provides insight into the thermal hazard of the ejecta. As shown in Figure 7, the maximum gas temperature for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery is 263.8 °C, while for the Na3V2(PO4)3 sodium-ion battery it is 173.9 °C. The higher gas temperature from the layered oxide sodium-ion battery aligns with its higher surface temperature, indicating greater thermal energy in the vented gases. Temperature衰减 with height is observed due to heat exchange and plume dispersion.
To quantify gas production, the constant-volume vessel tests are employed. The total gas yield is calculated using the ideal gas law:
$$ n_{\text{gas}} = \frac{P_{\text{gas}} V_{\text{container}}}{R T_{\text{gas}}} – \frac{P_0 V_{\text{container}}}{R T_0} $$
where \(n_{\text{gas}}\) is the molar amount of gas produced, \(P_{\text{gas}}\) and \(T_{\text{gas}}\) are the final pressure and temperature in the vessel, \(P_0\) and \(T_0\) are the initial pressure and temperature, \(V_{\text{container}} = 140\) L is the vessel volume, and \(R = 8.314\) J/(mol·K) is the gas constant. The volumetric gas yield \(V_{\text{gas}}\) at standard conditions can be derived as:
$$ V_{\text{gas}} = \frac{n R T_{\text{gas}}}{P_{\text{gas}}} $$
From the pressure curves in Figure 8, the stabilized pressure yields a gas volume of 326.6 L for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery and 294.8 L for the Na3V2(PO4)3 sodium-ion battery. The corresponding molar amounts are 14.58 mol and 13.16 mol, respectively. The maximum pressure rise rate, indicative of gas generation intensity, is 17.71 kPa/s for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery and 24.89 kPa/s for the Na3V2(PO4)3 sodium-ion battery, highlighting the more abrupt gas release in the latter sodium-ion battery type.
Gas composition analysis via GC-MS reveals that the vent gas from sodium-ion batteries is predominantly composed of CO2, H2, CO, and light hydrocarbons (CH4, C2H4, C2H6, C3H6, C3H8), which together account for over 95% of the total gas. The detailed compositions are summarized in Table 3. Notably, the sum of CO and CO2 constitutes about 50-60% of the gas mixture, which is higher than typical values for lithium iron phosphate batteries, suggesting a more oxidative decomposition pathway in these sodium-ion batteries.
| Gas Component | NaNi1/3Fe1/3Mn1/3O2 Sodium-Ion Battery | Na3V2(PO4)3 Sodium-Ion Battery |
|---|---|---|
| CO2 | 32.5% | 35.2% |
| H2 | 28.1% | 25.8% |
| CO | 22.4% | 20.1% |
| CH4 | 6.3% | 7.5% |
| C2H4 | 4.8% | 5.1% |
| C2H6 | 3.2% | 3.0% |
| C3H6 | 1.5% | 1.8% |
| C3H8 | 0.7% | 0.9% |
| Others | 0.5% | 0.6% |
The flammability of the gas mixture is assessed by calculating the lower explosion limit (LEL) using Le Chatelier’s law:
$$ \text{LEL}_{\text{mix}} = \frac{1}{\sum_{i=1}^{n} \frac{x_i}{\text{LEL}_i}} $$
where \(x_i\) is the volume fraction of component \(i\), \(\text{LEL}_i\) is its individual LEL, and \(n\) is the number of components. Using standard LEL values (e.g., H2: 4.0%, CO: 12.5%, CH4: 5.0%, etc.), the LEL for the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery gas mixture is computed as 7.5%, while for the Na3V2(PO4)3 sodium-ion battery it is 8.0%. These values are slightly higher than those for lithium iron phosphate batteries (typically 6-7.5%), indicating a moderate but still significant explosion risk for sodium-ion battery vent gases.
To integrate the multifaceted hazards, a thermal runaway hazard assessment model is developed using a radar chart that incorporates six key parameters: thermal runaway onset time (indicating susceptibility), maximum surface temperature (thermal severity), internal propagation rate (thermal propagation speed), peak gas temperature (gas thermal hazard), total gas volume (gas quantity hazard), and LEL (explosion hazard). Each parameter is normalized based on the experimental data, and the resulting profiles for both sodium-ion battery types are compared. As illustrated in Figure 10, the NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery exhibits a larger area in the radar chart, signifying a higher overall hazard level. Specifically, this sodium-ion battery shows shorter onset time, higher temperatures, faster propagation, larger gas volume, and lower LEL, collectively pointing to greater danger during thermal runaway events. In contrast, the Na3V2(PO4)3 sodium-ion battery demonstrates better thermal stability and lower gas-related risks, making it a safer choice for applications where safety is paramount.
The underlying mechanisms for the observed differences stem from the cathode material properties. The layered oxide cathode (NaNi1/3Fe1/3Mn1/3O2) in the sodium-ion battery has lower thermal stability, undergoing exothermic decomposition at lower temperatures and releasing more heat due to oxygen evolution and reactions with the electrolyte. The phosphate-based cathode (Na3V2(PO4)3) in the other sodium-ion battery possesses a more robust crystal structure, delaying decomposition and mitigating heat generation. Furthermore, the gas generation pathways differ: the layered oxide sodium-ion battery produces more CO and H2 via electrolyte reduction and cathode decomposition, while the phosphate sodium-ion battery generates relatively more CO2 from carbonate solvent oxidation. These compositional variations affect the flammability and explosion limits of the vent gases.
In practical terms, the findings emphasize that sodium-ion battery safety is highly dependent on cathode chemistry. For large-scale energy storage systems using sodium-ion batteries, designers must consider the trade-offs between energy density and safety. The NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery, while potentially offering higher energy, requires enhanced thermal management, robust venting design, and possibly flame-retardant additives to mitigate risks. The Na3V2(PO4)3 sodium-ion battery, with its superior thermal stability, may be preferable for stationary storage where safety margins are critical. Additionally, gas detection and ventilation systems should be tailored to the specific gas profiles and LEL values of the sodium-ion battery type deployed.
Future research directions include investigating the effects of state of charge, cycling aging, and module-level interactions on thermal runaway behavior in sodium-ion batteries. Moreover, the development of advanced electrolytes and coatings to suppress gas generation and enhance thermal stability in sodium-ion batteries is essential. Computational modeling coupled with the experimental data presented here can facilitate the prediction of thermal runaway propagation in sodium-ion battery packs, enabling proactive safety measures.
In conclusion, this experimental study provides a detailed characterization of thermal runaway in large-capacity sodium-ion batteries, highlighting the significant influence of cathode material on hazard severity. The NaNi1/3Fe1/3Mn1/3O2 sodium-ion battery exhibits higher thermal hazards, including lower onset temperature, higher peak temperatures, faster internal propagation, and greater gas emission with a lower explosion limit. The Na3V2(PO4)3 sodium-ion battery demonstrates better thermal stability and lower gas-related risks. These insights underscore the importance of material selection and system design in ensuring the safe adoption of sodium-ion battery technology for grid-scale energy storage. By quantifying key safety parameters, this work contributes to the foundational knowledge needed to develop effective risk mitigation strategies and safety standards for sodium-ion battery-based energy storage systems.