1. Introduction
Lithium-ion batteries, especially LiFePO4 batteries, have been widely used in various fields due to their high energy density and long cycle life. However, the safety issue during the application process remains a significant challenge. Thermal runaway is a critical problem that can lead to severe consequences, such as fires and explosions. Understanding the thermal runaway behavior and the influence of different factors is crucial for improving the safety and reliability of lithium-ion batteries.
In recent years, many studies have been conducted on the thermal runaway of lithium-ion batteries. However, most of the previous research focused on small-capacity batteries or specific cathode materials, and the influence of test atmosphere on the thermal runaway process of large-capacity LiFePO4 batteries has not been fully explored. This article aims to fill this gap by conducting a comprehensive study on the thermal runaway behavior of 120 Ah LiFePO4 batteries under different test atmospheres and analyzing the associated characteristics and influence laws.
2. Experimental Setup
2.1 Battery Samples
The experimental object is a square LiFePO4 battery with the following specifications:
| Parameter | Value |
|---|---|
| Cathode material | LiFePO4 |
| Anode material | Graphite |
| Electrolyte composition | (EC):(DMC)=1:1 |
| Cell capacity (Ah) | 120 |
| Dimensions (mm) | 174x170x48 |
| Initial mass (g) | 2860 |
| Nominal voltage (V) | 3.20 |
| Specific energy (Wh/kg) | 134.3 |
| Cut-off voltage lower limit (V) | 2.50 |
| Cut-off voltage upper limit (V) | 3.65 |
2.2 Experimental Apparatus
The experiments were carried out in a sealed pressure chamber with a volume of 82 L. The battery was heated laterally using a heating plate with a power of 952 W. Mica plates were attached to both sides of the battery, and an aluminum clamp was used to hold the battery tightly to reduce heat dissipation. Temperature sensors (K-type thermocouples) were placed at different positions on the battery surface and around the battery to monitor the temperature changes. A pressure sensor was installed inside the chamber to measure the pressure variation.
2.3 Experimental Procedure
The battery was charged to 100% SOC in a charge-discharge cycle tester and then placed in the pressure chamber for testing. The experiments were conducted in both inert (nitrogen) and air atmospheres, and each experimental condition was repeated twice. During the experiment, the temperature, pressure, and voltage of the battery were continuously monitored. After the experiment, the gas generated during thermal runaway was collected using a gas sampling bag and analyzed using a GC gas chromatograph.
3. Thermal Runaway Process Analysis
3.1 Battery Characteristic Temperature
The temperature changes of the battery during thermal runaway in an inert atmosphere are shown in Figure 2. The thermal runaway process can be divided into four stages:
- Stage I (0 – 904 s): Heating stage
All monitoring points show a continuous temperature increase. The temperature near the exhaust valve and the side center is the highest due to the proximity to the heating plate, while the back surface temperature is the lowest because of the small thermal conductivity of the battery core along the thickness direction. - Stage II (904 – 1290 s): Exhaust stage
At 904 s, the battery exhaust valve opens due to the evaporation of the electrolyte and the generation of thermal runaway gas. The temperature at the exhaust valve slightly decreases as the gas jet takes away some heat. The average temperature of the non-heating surface is 78.4 °C, and the temperature rise rate is 0.6 °C/s. - Stage III (1290 – 1750 s): Thermal runaway stage
The internal chemical reaction intensifies, and the temperature at each monitoring point rises rapidly. The voltage starts to drop at 1330 s and reaches 0 V at 1438 s. The battery spontaneously undergoes thermal runaway at 1290 s. The highest temperature is observed at the exhaust valve, reaching 162.4 °C. The back surface temperature is the lowest. During this stage, the battery experiences violent eruption, and the temperature at the exhaust valve shows a significant decrease. The maximum surface average temperature of 195.1 °C is reached at 1750 s, and the maximum temperature rise rate of 3.9 °C/s occurs at 1521 s. - Stage IV (after 1750 s): Cooling stage
The battery temperature gradually decreases.
The characteristic temperature data in different atmospheres are summarized in Table 1.
| Test Atmosphere | Valve Opening Temperature (°C) | Thermal Runaway Starting Temperature (°C) | Peak Temperature (°C) | Maximum Temperature Rise Rate (°C/s) | Highest Ambient Temperature (°C) |
|---|---|---|---|---|---|
| Inert Atmosphere | 117.5 | 99.5 | 218.0 | 3.9 | 132.0 |
| Air Atmosphere | 122.3 | 102.3 | 256.3 | 4.1 | 133.0 |
| Characteristic Difference (%) | 4.1 | 2.8 | 17.6 | 5.1 | 0.8 |
It can be seen that the air atmosphere has a significant impact on the peak temperature and maximum temperature rise rate during the thermal runaway stage.
3.2 Ambient Temperature
The ambient temperature changes around the battery during thermal runaway are shown in Figure 3. When the valve opens, the ambient temperature shows a small 突变 due to the high-temperature gas jet. During the thermal runaway stage, the ambient temperature reaches a peak, with the highest temperature of 131.7 °C appearing at 20 cm to the left of the battery. The ambient temperature above the battery decreases with increasing height, indicating that the danger to the upper environment decreases. The temperature distribution in different directions is summarized in Table 2.
| Position | Temperature (°C) |
|---|---|
| 10 cm above the battery | 82.8 (at 1357 s) |
| 20 cm above the battery | 86.0 (at 1362 s) |
| 30 cm above the battery | 78.2 (at 1354 s) |
| 20 cm to the left of the battery | 131.7 |
| 20 cm to the right of the battery | 96.3 |
The divergent exhaust behavior of the battery during thermal runaway causes the peak temperature to appear on the side of the battery, highlighting the importance of considering the impact on the side environment.
3.3 Gas Production Kinetics
The pressure and gas production changes in the sealed chamber during the thermal runaway of the battery in an inert atmosphere are shown in Figure 4. When the exhaust valve opens, the chamber pressure increases slightly, and the pressure rise rate is 0.18 kPa/s. During thermal runaway, the pressure reaches a maximum value of 215.2 kPa at 1458 s, and the maximum pressure rise rate is 1.3 kPa/s. The gas production and exhaust rate also show corresponding changes. The gas production reaches a peak of 35.5 L at 1459 s, and the exhaust rate reaches a maximum of 19.7 L/s at 1326 s. The data in different atmospheres are compared in Table 3.
| Test Atmosphere | Gas Production (L) | Exhaust Rate (L/s) |
|---|---|---|
| Inert Atmosphere | 35.5 | 19.7 |
| Air Atmosphere | 38.4 | – |
The air atmosphere increases the gas production during thermal runaway, which may increase the risk of explosion.
3.4 Gas Composition Analysis
The gas composition produced during the thermal runaway of the battery in an inert atmosphere is analyzed, and the volume fraction of each component is shown in Figure 5. The main gas components are H2 (52.8%), CO2 (26.5%), CO (7.4%), CH4 (6.2%), and C2H4 (5.0%). The sources of these gases are as follows:
- H2: Generated by the reaction between the binder and lithium ions at high temperatures. When the core temperature reaches about 230 °C, the graphite anode may shed, and the metal lithium reacts with the binder (PVDF and CMC) to produce H2.
- CO: Produced by the decomposition of the SEI film and the reduction reaction of the electrolyte with the anode lithium ions at high temperatures.
- CH4: Generated by the reaction between the binder reaction product (H2) and the electrolyte (DMC) at high temperatures.
- CO2: Comes from the decomposition of the SEI film, the reaction between the SEI film and the active material, and the reaction between the lithium carbonate in the positive electrode and hydrofluoric acid.
- C2H4: Generated by the reaction between the lithium embedded in the graphite anode and the electrolyte (EC) after the anode collapses and the decomposition of the SEI film.
The gas composition in different atmospheres is compared in Figure 6. Although the types of gas components are the same in both atmospheres, the proportion of each component is slightly different. In the air atmosphere, the proportion of H2 is lower, while the proportions of CH4 and C2H4 are higher. This is because the higher reaction temperature in the air atmosphere promotes more reactions that consume H2 and generate CH4 and C2H4. The flammability limits of the gas mixture in different atmospheres are calculated, and both show a high explosion risk.
4. Influence of Test Atmosphere
4.1 Characteristic Temperature Difference
As shown in Table 1, the air atmosphere has a significant impact on the peak temperature and maximum temperature rise rate during the thermal runaway stage. Before the battery undergoes thermal runaway, the influence of the test atmosphere on the characteristic temperature is not significant. However, during the thermal runaway stage, the peak temperature in the air atmosphere is 17.6% higher than that in the inert atmosphere, and the maximum temperature rise rate is 5.1% higher. This is because when the exhaust valve opens, the active materials inside the battery start to react with the air. Although the reaction is not significant at low temperatures, it becomes severe during the thermal runaway stage, resulting in a higher peak temperature.
4.2 Characteristic Time Difference
The characteristic time differences in different atmospheres are shown in Table 3. The opening time and exhaust duration of the valve are not significantly affected by the test atmosphere. However, the thermal runaway starting time is earlier and the ending time is later in the air atmosphere, resulting in a 13.9% longer thermal runaway duration. This is because the reaction between the active materials and air in the air atmosphere accelerates the temperature rise and prolongs the reaction time during thermal runaway.
4.3 Gas Production Characteristic Difference
As shown in Figure 7 and Table 3, the gas production in the air atmosphere is 8.2% higher than that in the inert atmosphere. This is due to the more 充分 reaction between the oxygen in the air and the active materials inside the battery during the thermal runaway stage. Although the types of gas components are the same in both atmospheres, the proportion of each component is slightly different. The air atmosphere reduces the proportion of H2 and increases the proportions of CH4 and C2H4. The flammability limits of the gas mixture in both atmospheres indicate a high explosion risk.
5. Conclusion
In this study, the thermal runaway behavior of 120 Ah LiFePO4 batteries under different test atmospheres was investigated. The results show that the thermal runaway process of the battery exhibits multi-stage characteristics, including heating, valve opening, thermal runaway, and cooling. The air atmosphere has a significant impact on the thermal runaway process, increasing the peak temperature, prolonging the thermal runaway duration, and increasing the gas production. The main gas components produced during thermal runaway are H2, CO2, CO, CH4, and C2H4, and the gas mixture has a high explosion risk. Understanding these characteristics and influence laws is crucial for developing effective safety measures and early warning systems for lithium-ion batteries, especially in applications such as energy storage systems where large-capacity batteries are used. Future research can focus on further exploring the reaction mechanisms and developing more accurate models to predict and prevent thermal runaway accidents.