The proliferation of electric vehicles has brought the thermal safety of power batteries into sharp focus. Statistics from leading research laboratories, analyzing numerous publicly reported incidents, underscore the severe consequences of thermal runaway in li ion battery systems, including substantial property damage and personal injury. To meet the escalating demand for extended driving range, national policies continuously push for higher specific energy in traction batteries. However, this pursuit often comes with a critical trade-off: as the specific energy of a li ion battery increases, its thermal stability typically degrades. Consequently, suppressing thermal runaway and mitigating its hazards have become paramount research priorities in the development of safe, high-energy-density li ion battery technologies.
The initiation of thermal runaway in a li ion battery is generally attributed to three primary abuse conditions: mechanical, electrical, and thermal abuse. Mechanical abuse involves physical deformation of the cell due to incidents such as crushing, impact, or penetration, leading to internal short circuits and subsequent uncontrolled heating. Electrical abuse stems from management or component failures, resulting in conditions like overcharging, over-discharging, or internal and external short circuits. Thermal abuse occurs when the battery is subjected to excessive external temperatures, triggering exothermic decomposition reactions. Crucially, when one cell within a module or pack undergoes thermal runaway, the massive heat release can propagate to adjacent cells, triggering a chain reaction known as thermal propagation or thermal runaway propagation. This phenomenon can lead to the complete failure of the entire energy storage system.
Among commercial cell chemistries, nickel-manganese-cobalt (NMC) based li ion battery cells are widely used due to their high energy density. Yet, this very attribute is coupled with higher intrinsic chemical reactivity, making the management of their safety under abuse conditions a significant challenge. This study focuses on the thermal behavior of a commercial NMC pouch li ion battery module subjected to the most severe form of mechanical abuse testing: nail penetration. We investigate the thermal propagation characteristics at the module level. Furthermore, a computational thermal propagation model is developed and rigorously calibrated against experimental data. The validated model serves as a powerful tool for rapidly assessing thermal runaway risks, enabling the design of effective mitigation strategies, and significantly reducing the reliance on costly and hazardous physical tests during the development phase of li ion battery modules and packs.
Experimental Investigation of Nail-Induced Thermal Propagation
The experimental study was designed to replicate a worst-case mechanical abuse scenario within a simplified module configuration. The test specimen consisted of six commercial 53 Ah NMC (Lithium Nickel Manganese Cobalt Oxide) pouch li ion battery cells arranged in a stack. Each cell was constrained with a specified preload on its large surface area to simulate typical module assembly conditions. A standardized nail, with a diameter of 1.5 mm and a conical tip angle of 30°, was used as the penetration tool. The nail was driven into the geometric center of the first cell’s large face at a constant velocity of 80 mm/s to a depth of 10 mm, ensuring a severe internal short circuit.
To capture the spatiotemporal evolution of temperature during the thermal runaway and subsequent propagation events, an array of thermocouples was strategically installed. On the large face of each of the six cells, five thermocouples (labeled T1 through T5) were positioned. The central thermocouple (T5) was placed at the face center, while the others were distributed towards the cell’s edges and tabs. This arrangement allowed for detailed mapping of the temperature front as it moved from the triggered cell through the module. Voltage of each cell was also monitored throughout the test to correlate electrical failure with thermal events. The experimental setup and thermocouple layout are conceptualized in the schematic below, detailing the cell stack, nail position, and sensor placement.
| Parameter | Specification |
|---|---|
| Cell Type | Commercial NMC Pouch Li Ion Battery |
| Capacity | 53 Ah |
| Module Configuration | 6 cells in series/parallel stack |
| Preload Force | 1,083 N per cell face |
| Nail Diameter | 1.5 mm |
| Nail Tip Angle | 30° |
| Penetration Speed | 80 mm/s |
| Penetration Depth | 10 mm |
| Trigger Location | Center of first cell’s large face |
Analysis of Thermal Propagation Behavior
The nail penetration successfully induced thermal runaway in the first cell, which then propagated sequentially through the entire six-cell module. The temperature data revealed critical insights into the dynamics of heat transfer during this failure mode. Analysis of the temperature profiles from the central thermocouples (T5) on cells 2 through 6 shows a sequential ignition with distinct time delays, as summarized in the table below.
| Cell Number | Approx. Time to Thermal Runaway After Trigger (s) | Peak Temperature at T5 (°C) |
|---|---|---|
| 1 (Triggered) | 0 (Reference) | >900 |
| 2 | ~35 | >900 |
| 3 | ~55 | >900 |
| 4 | ~80 | >900 |
| 5 | ~105 | >900 |
| 6 | ~130 | >900 |
A more detailed examination of the temperature distribution on each cell’s face provided the most significant finding. For the first few cells (Cells 2, 3, and 4), the thermocouple at the center position (T5) registered a rapid temperature rise first, followed by the thermocouples closer to the edges (T1, T2, T3, T4). This pattern indicates that the leading edge of the propagating thermal front was not planar but had an arched or dome-like shape. The heat from the runaway cell was transferred more effectively through the center of the adjacent cell’s face, creating a hot spot that then spread laterally towards the tabs and edges.
Interestingly, this pattern shifted for the last two cells in the module (Cells 5 and 6). For these cells, thermocouples near the positive tab side (T3) registered the initial temperature increase slightly before the central thermocouple (T5). This suggests a change in the dominant heat transfer mechanism or propagation front geometry as the event progressed through the stack. A plausible explanation is the directional venting of hot gases and ejecta from the preceding runaway cells. As cells fail, they often vent from a specific location (e.g., near the tabs). For this particular pouch li ion battery, venting may have occurred preferentially towards the positive tab side in the later stages of propagation. This directed jet of hot material could then impinge on the adjacent cell’s surface near the tab area, locally heating it faster than the conductive heat transfer through the cell stack’s center. This observation highlights that thermal propagation in a li ion battery module is not solely a conductive process but can be significantly influenced by convective and radiative heat transfer from ejected materials.
Development and Validation of a Thermal Propagation Model
Model Formulation
To enable predictive analysis and design optimization, a three-dimensional finite element thermal model of the li ion battery module was developed. The geometry included the six pouch cells, the mica insulation sheets placed on the outer surfaces of the end cells, and the nail penetrating the first cell. The core of the model is governed by the transient heat conduction equation with an internal heat generation source term representing the energy released during thermal runaway of each li ion battery:
$$ \rho C_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \dot{q}_{gen} $$
where:
– $\rho$ is the density of the material (cell, insulation).
– $C_p$ is the specific heat capacity.
– $T$ is the temperature.
– $t$ is time.
– $k$ is the thermal conductivity tensor (which can be anisotropic, especially for a pouch li ion battery where in-plane and through-plane conductivities differ).
– $\dot{q}_{gen}$ is the volumetric heat generation rate ($W/m^3$).
The total heat released $Q_{TR}$ by a single li ion battery during thermal runaway is the integral of the generation rate over the cell volume $V$ and the duration of the event:
$$ Q_{TR} = \int_{t} \int_{V} \dot{q}_{gen} \, dV \, dt $$
This heat release $Q_{TR}$ was modeled as a time-dependent power curve calibrated against experimental data. The model accounted for key thermal properties of the cell components (jellyroll, casing, tabs) and the mica insulation. Thermal contact resistances between adjacent cells and between cells and insulation were also considered, as they critically impact the rate of inter-cell heat transfer in a li ion battery stack.
Model Calibration and Results Verification
The thermal propagation model was rigorously calibrated using the experimental data obtained from the nail penetration test. The calibration process involved adjusting parameters such as the exact timing, magnitude, and shape of the heat release function $\dot{q}_{gen}(t)$ for a single cell, as well as the effective thermal contact resistances between cell surfaces, to match the observed temperatures and propagation delays.
The calibrated model demonstrated a high degree of accuracy in replicating the experimental behavior. As shown in the comparison below, the simulated temperature profiles at the central points (T5) of cells 2 through 6 closely match the experimental measurements in terms of both the time-to-thermal-runaway and the peak temperatures reached.
| Cell Number | Experimental TR Delay (s) | Simulated TR Delay (s) | Absolute Error (s) | Accuracy in Timing |
|---|---|---|---|---|
| 2 | ~35 | ~33 | 2 | 94.3% |
| 3 | ~55 | ~52 | 3 | 94.5% |
| 4 | ~80 | ~75 | 5 | 93.7% |
| 5 | ~105 | ~99 | 6 | 94.3% |
| 6 | ~130 | ~119 | 11 | 91.5% |
Furthermore, the model successfully captured the key spatial phenomenon: the arched propagation front. The simulated temperature contours clearly showed that for the initial cells in the propagation sequence, the highest temperature region developed at the center of the cell face before expanding outward. The simulation also indicated that for cells further down the chain, the influence of directional heating could alter this pattern, consistent with the experimental observation for the last two cells. The validated model thus serves as a high-fidelity digital twin for the thermal propagation behavior of this specific li ion battery module under nail abuse.
Discussion on Propagation Mechanics and Model Utility
The experimental and simulation results collectively paint a clear picture of thermal propagation dynamics in a closely packed pouch li ion battery module. The primary mechanism for heat transfer from a runaway cell to its immediate neighbor is conduction through the contact area. Because the nail triggers the first cell at its center, this becomes the initial epicenter of heat generation. This heat conducts into the adjacent cell, primarily through the central region of their shared interface, naturally establishing a dome-shaped temperature field that then spreads laterally. This explains the observed “center-first” heating pattern in the early stages of propagation for this module configuration.
The shift in the initiation location for later cells underscores the multi-physics nature of a full-scale li ion battery thermal runaway event. As cells fail violently, they release not just heat but also hot gases, flames, and solid particulates. The path and intensity of this ejecta can become a significant secondary heating mechanism, potentially overwhelming the purely conductive heat transfer, especially if venting is directed. A robust thermal propagation model for a li ion battery system must therefore be capable of integrating, or at least accounting for, these convective/radiative effects to be fully predictive across all scenarios.
The high accuracy (exceeding 93.7% for timing in most cases) of the developed conduction-centric model for this specific test validates its core thermal parameters and assumptions. Its primary utility lies in the rapid, cost-effective, and safe exploration of design parameters. Engineers can use such a model to:
- Evaluate Thermal Barriers: Virtually test different interstitial materials (e.g., aerogels, phase change materials, enhanced insulation) by modifying their thermal properties (k, ρ, C_p) in the model to assess their effectiveness in delaying or stopping propagation.
- Optimize Module Layout: Study the impact of cell spacing, stack size, and cooling plate integration on propagation limits.
- Perform Risk Assessment: Simulate propagation from different trigger locations (corner, tab side) to identify worst-case scenarios and vulnerable areas in the li ion battery pack design.
- Reduce Physical Testing: Significantly cut down the number of destructive and expensive nail penetration or thermal propagation tests required during the design validation phase.
Conclusion
This study provides a detailed investigation into the thermal runaway propagation behavior of a commercial NMC pouch li ion battery module subjected to nail penetration abuse. The key findings are:
- Thermal propagation under these conditions proceeds via a domed or arched frontal surface through the cell stack, with the center of the adjacent cell’s face heating initially before the edges, due to conductive heat transfer from the central trigger point.
- In later stages of propagation, directional effects such as venting of hot ejecta can alter the heat transfer mode, potentially causing localized heating near cell tabs before the center.
- A three-dimensional finite element thermal model was successfully developed and calibrated against experimental data. The model demonstrates high accuracy in predicting both the timing of sequential thermal runaway events and the spatial temperature distribution, validating its parameters and assumptions.
- The established model serves as a powerful and efficient engineering tool for analyzing thermal risks and evaluating mitigation strategies in li ion battery module and pack design, offering a substantial reduction in the reliance on hazardous and costly physical abuse testing.
This work summarizes the specific thermal propagation characteristics following nail penetration for a commercial ternary li ion battery module and provides a validated simulation methodology. The insights and the modeling framework contribute valuable knowledge and tools for advancing the safety design of high-energy-density li ion battery systems against mechanical abuse.