In the field of energy storage and electric vehicles, the safety of lithium-ion batteries remains a critical concern, particularly under thermal runaway conditions. Thermal runaway in a lithium-ion battery involves complex exothermic reactions and gas generation, which can lead to catastrophic failures if not properly managed. While previous studies have focused on thermal behavior, the gas generation aspects, especially for large-capacity prismatic cells, are less explored. This study aims to investigate the heat and gas generation characteristics of large-capacity lithium-ion batteries after thermal runaway, using advanced experimental techniques. The findings are intended to support the development of comprehensive simulation models that integrate both heat and gas production processes for accurate safety assessments.
The lithium-ion battery used in this research is a commercial 40 A·h prismatic ternary (NCM/C) cell, with key parameters summarized in Table 1. This battery type is widely employed in electric vehicles and energy storage systems, making it a relevant subject for safety analysis. The experiments were conducted in two phases: first, to measure the thermal properties and heat generation during thermal runaway using an Accelerating Rate Calorimeter (ARC), and second, to evaluate gas generation using a custom-built large-capacity lithium-ion battery thermal runaway gas testing apparatus. The integration of these data sets allows for a holistic understanding of the thermal runaway phenomenon.
To begin, the specific heat capacity of the lithium-ion battery was determined as a foundational parameter for heat calculations. The ARC provides a near-adiabatic environment, minimizing heat exchange with the surroundings. In this test, two battery samples were sandwiched with a heating element, and a constant power was applied to heat the cells uniformly. The temperature rise was monitored, and the average specific heat capacity over the 30–60°C range was calculated using the formula:
$$ c_{av} = \frac{P}{m} \cdot \frac{\partial t}{\partial T} $$
where \( P = 6 \, \text{W} \) is the heating power, \( m = 1.65 \, \text{kg} \) is the total mass of the two batteries, \( \partial T = 30 \, ^\circ\text{C} \) is the temperature change, and \( \partial t = 10230 \, \text{s} \) is the time duration. The calculated average specific heat capacity was \( c_{av} = 1.242 \, \text{kJ/(kg·K)} \). This value is essential for subsequent heat generation analyses, as it influences the energy balance during thermal runaway events in lithium-ion batteries.
| Parameter | Value |
|---|---|
| Positive/Negative Electrode Material | NCM/C |
| Dimensions (mm) | 27.00 × 148.00 × 91.00 |
| Mass (kg) | 0.82 |
| Rated Capacity (A·h) | 40.00 |
| Rated Voltage (V) | 3.65 |
| Charge Cut-off Voltage (V) | 4.20 |
| Discharge Cut-off Voltage (V) | 2.80 |
Following the specific heat measurement, adiabatic thermal runaway tests were performed on a fully charged (SOC = 100%) lithium-ion battery using the ARC. The procedure involved heating the battery from ambient temperature to 60°C, followed by stepwise heating at 5°C increments. When the self-heating rate exceeded 0.02°C/min, the battery was considered to have initiated self-heating. The ARC then maintained adiabatic conditions, allowing all generated heat to contribute to the battery’s temperature rise. The thermal runaway event was identified when the temperature rise rate reached or exceeded 1°C/s along with a voltage drop of over 25%. The results, as shown in Figure 2 (temperature vs. time curve), indicated a self-heating onset at 75°C, a thermal runaway temperature of 165°C, and a peak temperature of 668°C. The mass of the battery decreased from 0.825 kg to 0.537 kg after thermal runaway, indicating substantial material ejection.
The total heat released during thermal runaway was estimated by considering the heat retained in the battery body and the heat carried away by ejected materials. Assuming the ejected materials and gases reached the peak temperature of 668°C, the heat quantities were computed using:
$$ Q = c \cdot m \cdot \Delta T $$
where \( Q \) is the heat, \( c \) is the specific heat capacity, \( m \) is the mass, and \( \Delta T \) is the temperature change. The total heat generation was 587.07 kJ, with 204.94 kJ dissipated via ejecta and 382.13 kJ retained in the battery residue. These values are crucial for modeling the energy distribution during thermal runaway in lithium-ion batteries.
| Parameter | Value |
|---|---|
| Self-Heating Onset Temperature (°C) | 75 |
| Thermal Runaway Temperature (°C) | 165 |
| Peak Temperature (°C) | 668 |
| Initial Mass (kg) | 0.825 |
| Final Mass (kg) | 0.537 |
| Total Heat Generation (kJ) | 587.07 |
| Heat Carried by Ejecta (kJ) | 204.94 |
| Heat Retained in Battery (kJ) | 382.13 |
Moving to gas generation studies, a specialized apparatus was designed to withstand the high pressures and temperatures associated with large-capacity lithium-ion battery thermal runaway. The system consists of a thermal runaway chamber, a gas collection tank, vacuum pumps, and various sensors for pressure, temperature, voltage, and gas composition. The chamber is made of 2520 stainless steel with an internal volume of approximately 77 L, capable of resisting pressures up to 12 MPa and temperatures up to 1300°C. The workflow includes different modes such as air mode, gas replacement mode, and low-pressure mode, allowing for flexible testing under various conditions. Prior to actual tests, the apparatus was validated for sealing integrity and repeatability. Pressure and temperature curves from preliminary tests confirmed stable post-thermal runaway conditions, and repeated trials on identical batteries showed consistent peak pressures around 790 kPa, demonstrating reliable performance.
For gas generation evaluation, three identical lithium-ion batteries were arranged in parallel within the chamber, with a heating element triggering thermal runaway in one cell, which then propagated to the others. This setup minimizes external heat input variations and reduces individual cell discrepancies. The pressure and temperature inside the chamber were recorded during the thermal runaway events. As shown in Figure 10, three distinct pressure spikes corresponded to the sequential thermal runaways of the batteries, with gas ejection durations averaging about 7 seconds per cell. The total gas volume produced was calculated using the ideal gas law:
$$ PV = nRT $$
At a stable point after thermal runaway (t = 75 s), the absolute pressure \( P_a = 266 \, \text{kPa} \) (165 kPa gauge + 101 kPa atmospheric) and temperature \( T_a = 377.15 \, \text{K} \) were used. The volume of gas at standard temperature and pressure (STP) was derived as:
$$ V_0 = \frac{T_0 P_a}{T_a P_0} V_a $$
where \( V_0 \) is the gas volume at STP, \( T_0 = 298.15 \, \text{K} \), \( P_0 = 101 \, \text{kPa} \), and \( V_a = 77 \, \text{L} \). The calculated \( V_0 \) was 160 L. Accounting for the initial oxygen in the chamber (21% of 77 L) that reacted during thermal runaway, the net gas generated by the batteries was:
$$ V = V_0 – V_a + 0.21 V_a $$
resulting in \( V = 99 \, \text{L} \). Thus, each lithium-ion battery produced approximately 33 L of gas at STP. The gas composition was analyzed using gas chromatography, revealing a mix of CO₂, CO, H₂, CH₄, and C₂–C₄ hydrocarbons, as summarized in Table 3. CO₂ constituted about 45% of the total, while CO and H₂ together exceeded 40%. These gases originate from decomposition reactions of electrode materials, electrolytes, and binders under high-temperature conditions. For instance, carbonaceous materials oxidize to form CO₂ and CO, while electrolyte solvents react with lithium to produce hydrocarbons. The high proportion of CO₂ and CO aligns with previous studies on lithium-ion battery thermal runaway gas emissions.
| Gas Component | Mass Fraction (%) |
|---|---|
| CO₂ | 45 |
| CO | 25 |
| H₂ | 18 |
| CH₄ | 7 |
| C₂–C₄ Hydrocarbons | 5 |
The integration of heat and gas generation data is pivotal for advancing thermal runaway models. Traditional simulations often focus solely on heat transfer, but incorporating gas dynamics can improve accuracy, especially for large-capacity lithium-ion batteries where gas ejection influences internal pressure and thermal diffusion. The measured parameters, such as specific heat capacity, heat release quantities, gas volumes, and ejection times, serve as inputs for model calibration. For example, the heat generation rate can be modeled using Arrhenius-type equations based on temperature-dependent reaction kinetics, while gas production can be linked to mass loss and reaction stoichiometry. A proposed framework for a coupled model might include energy balance equations like:
$$ \frac{dQ}{dt} = \sum_i A_i \exp\left(-\frac{E_i}{RT}\right) + \dot{m}_g \cdot h_g $$
where \( \frac{dQ}{dt} \) is the heat generation rate, \( A_i \) and \( E_i \) are pre-exponential factors and activation energies for reactions, \( \dot{m}_g \) is the gas mass flow rate, and \( h_g \) is the enthalpy of gas ejection. Similarly, gas pressure buildup can be described by:
$$ \frac{dP}{dt} = \frac{RT}{V} \frac{dn}{dt} – \frac{P}{V} \frac{dV}{dt} $$
considering gas generation rate \( \frac{dn}{dt} \) and volume changes. These equations, when parameterized with experimental data, can simulate thermal runaway propagation in battery packs, aiding in safety design for lithium-ion battery systems.
Furthermore, the implications of gas generation extend beyond modeling. In practical applications, such as electric vehicles or grid storage, the sudden release of flammable and toxic gases poses fire and explosion hazards. The presence of CO and H₂, which are highly combustible, requires robust ventilation and gas detection systems. Additionally, the pressure rise from gas generation can mechanically stress battery enclosures, potentially leading to rupture. Therefore, understanding these aspects is essential for developing mitigation strategies, such as pressure relief devices or advanced thermal management systems. Future research could explore the effects of different state-of-charge levels, cycling history, or environmental conditions on gas generation in lithium-ion batteries.
In conclusion, this study provides comprehensive experimental data on the heat and gas generation characteristics of large-capacity lithium-ion batteries during thermal runaway. The 40 A·h prismatic cell exhibited a specific heat capacity of 1.242 kJ/(kg·K), with thermal runaway initiating at 75°C and peaking at 668°C, releasing 587.07 kJ of heat. Gas generation amounted to 33 L per battery at STP, with ejection lasting around 7 seconds, and the gas composition was dominated by CO₂ and CO. These findings enhance the understanding of thermal runaway mechanisms and offer valuable inputs for simulation models that combine thermal and gas dynamics. As lithium-ion batteries continue to power modern technology, such insights are critical for improving safety standards and preventing catastrophic failures. Ongoing work will focus on refining these models and extending tests to other battery chemistries and formats to build a more generalized safety framework.