3 Test results
3.1 Measurement and Analysis of Battery Component Dimensions
Measure the dimensions of key components and sub components of LFPB using a digital micrometer (for thickness measurement), digital vernier caliper (for width measurement), and tape measure (for length measurement). Except for the 32650 battery which has an additional central pin (for exhaust in case of short circuit and diaphragm burnout), all other components are the same. The safety valve assembly includes a positive temperature coefficient (PTC) device, a plate with a safety vent, a positive terminal plate, and a gasket seal. The core (including anode, cathode, and two-layer polymer porous membrane separator) is a key part of energy storage. The anode active material graphite is deposited on the surface of the copper current collector, and the cathode active material LFP coating is on the surface of the aluminum current collector. There is an ion permeable separator between the electrodes.
The battery capacity depends on the volume of the electrode active materials (graphite on copper foil, LFP on aluminum foil, and the total volume of both). The LFP volume of the 32650 battery is the highest, the LFP to graphite ratio of the 26650 battery is the highest (5.85), and the 18650 battery is the lowest (0.61). The 26650 battery uses the least amount of copper but has a higher capacity, indicating that the battery capacity not only depends on the LFP amount, but also on the LFP to graphite ratio. This is helpful for the development of low copper LFPB. Anatomical analysis found that the volume of active materials is directly related to capacity, affecting battery size and failure behavior.
3.2. Observation and Analysis of Failure Behavior in Different Tests
Smoke release and chemical reaction rate: In horizontal and vertical crush and nail penetration test, smoke (containing toxic lithium hexafluorophosphate, which decomposes into toxic hydrogen fluoride when exposed to water) is released when the battery fails. Horizontal compression releases smoke quickly, and the chemical reaction rate is high due to the participation of the electrode active material in short-circuit reactions along its entire length; Longitudinal compression results in slow smoke release, and the electrode contact area is small due to the bending and buckling of the core; The smoke release rate of the spike test is lower due to the small short-circuit area. Smoke release and short circuit occurrence are delayed by 1-3 seconds in horizontal and spike tests, and 5-8 seconds in vertical compression. Smoke release delay increases the risk of explosion. The test follows safety procedures (good ventilation, wearing FFP3 mask) and should avoid ventilation blockage (horizontal and spike tests reduce the risk of explosion due to ventilation and spike holes, while vertical compression increases the risk of ventilation blockage due to machine crossbeams).
Failure load and temperature changes
Lateral compression: The failure loads of 18650, 22650, 26650, and 32650 batteries are 44kN, 31.88kN, 12.78kN, and 6.68kN, respectively. The smaller the size, the greater the failure load, and the higher the degree of core compaction and temperature. The load displacement curve is divided into four stages (compression of battery terminals and shell, start of core compression, compression and bending of core causing short circuit, and complete failure of battery after severe short circuit). This study focuses on the first three stages, where the initial voltage of the short circuit decreases (soft short circuit), and further loading will result in a severe hard short circuit. The first voltage drop can serve as a warning to control thermal runaway.
Longitudinal compression: The four types of battery failure loads mentioned above are 2.92kN, 7.76kN, 11.44kN, and 31.24kN, respectively. The smaller the size, the smaller the failure load. The load acts along the axial direction, causing the core to bend and bend. The electrode contact area of small-sized batteries is small. The temperature rise of 18650 batteries is the lowest during short circuit, while that of 32650 batteries is the highest. The sudden drop in load during the lateral compression stage III of the 32650 battery is related to the compression of the central hollow pin, which provides buckling strength during longitudinal compression, resulting in a higher failure load. This indicates that the battery needs to optimize its size to control the risk of thermal runaway.
Nail test: The 32650 battery has the highest failure load, and the smaller the battery size, the smaller the nail depth required for internal short circuit to occur.
Temperature rise and risk of thermal runaway: The safe temperature range for lithium-ion batteries is 20-60 ℃. During testing, loading causes the surface temperature to rise due to internal short circuit heat release reactions and Joule heating. Both soft and hard short circuits are caused by the failure of the battery separator, allowing charges to flow through a zero resistance path. The soft short circuit electrode layer has less contact, resulting in less voltage drop and temperature rise, while the hard short circuit electrode layer has more contact, and the heat is likely to cause thermal runaway, fire or explosion. The temperature rise should be controlled within a safe range. Soft short circuits are thermal runaway triggering events, and the test results can be used to develop early warning systems. When compressed horizontally, the temperature of 18650 batteries rises by about 38.75 ℃, which is higher than the optimal range (15-35 ℃). The delay in temperature rise indicates a low heat generation rate and is important for thermal management; During longitudinal compression, the temperature of 18650 batteries rises earliest and is delayed as the battery diameter increases; In the spike test, the 32650 battery had the highest failure load, and the short-circuit temperatures of 18650 and 22650 were similar. The higher sizes of 26650 and 32650 had lower short-circuit temperatures, with 18650 being the earliest to short-circuit.
3.3 Appearance analysis of failed samples
The failure samples after mechanical abuse testing showed that the battery used in the spike test had nail holes. The 18650 battery’s longitudinal compression test formed a ring due to buckling, indicating ISC reasons. The 26650 battery’s longitudinal compression test caused a mild explosion due to sudden rupture and tearing of the shell (due to the ventilation opening being blocked by the crossbeam). During the test, high-temperature tape was used to avoid welding line failure or disconnection, and subsequent tests drilled holes in the tape to maintain ventilation.
4 Summary
An anatomical analysis of the new battery revealed that the LFPB 32650 battery has an additional central pin, which serves as a channel for gas/smoke to flow from the negative electrode to the positive electrode during thermal runaway, providing an extra safety measure. The main conclusions of this study regarding the size dependence of failure behavior of LFPB with different sizes under mechanical abuse conditions are as follows:
Lateral compression test: As the diameter of the battery increases, the failure load decreases. The lateral compression failure loads of 18650, 22650, 26650, and 32650 batteries are 44kN, 31.88kN, 12.78kN, and 6.68kN, respectively.
Longitudinal compression test: The larger the battery size, the greater the failure load. The longitudinal compression failure loads of 18650, 22650, 26650, and 32650 batteries are 2.92kN, 7.76kN, 11.44kN, and 31.24kN, respectively.
Spike test: The 32650 battery failure load is the highest among the considered batteries. Meanwhile, in longitudinal and spike tests, an increase in battery diameter can delay the occurrence of internal short circuits, while in transverse compression tests, larger sized batteries experience short circuits earliest.
Conclusion on temperature changes during short circuit: The temperature rise during short circuit varies among batteries of different sizes. When compressed longitudinally, the temperature rise of the 32650 battery can reach up to 64.4 ℃, while when compressed laterally, it can reach as low as 29.5 ℃. In the longitudinal compression test, the temperature rise of the 32650 battery during short circuit (higher than the ambient value) was as high as 36.4 ℃. In contrast, the 26650 battery exhibited more balanced thermal behavior under various mechanical abuse conditions.
Conclusion on Smoke Release and Explosion Risk: In horizontal and spike tests, the delay between smoke release and short circuit occurrence is up to 3 seconds, and in vertical compression tests, it is up to 8 seconds. Delay in gas/smoke release increases the risk of gas accumulation and explosion, such as in the 26650 battery test where Kapton tape fixed thermocouples blocking ventilation openings increased the risk of explosion.
These test results are of great significance for the development of battery testing standards and early failure detection systems for battery packs, which can help improve the safety of electric vehicles and provide important basis for evaluating and improving battery safety performance. They can guide the rational selection and use of batteries of different sizes in battery design and application, in order to reduce the risk of failure.