Benefits
Events
Products & Programs
Author:
Elizabeth Favreau
Marketing Writing Team Lead
The need to reduce emissions from the transportation sector has spawned a new era of evolution and development in the automotive industry. Many countries around the world have identified electric vehicles as a crucial piece of the decarbonization puzzle, and as with any emerging technology (or, more precisely in the case of electric vehicles, reemerging), safety is a primary concern.
While statistically the safety of electric vehicles is on par with conventional powertrains, battery thermal runaway and thermal propagation have been thrust into the spotlight as a potential hazard. In the worst-case scenarios, thermal propagation can lead to battery fires or explosions, which pose a threat to vehicle occupants and can release toxic gases into the environment.
Convergent Science recently teamed up with IAV to take on the problem of thermal runaway propagation. IAV is an international company based in Berlin, Germany, that has been developing technical solutions for the automotive industry for over forty years. They are dedicated to providing the best expertise and methodologies to their customers to help them tackle challenging engineering problems, such as thermal propagation in electric vehicle batteries.
“There are many reasons why it’s important to study thermal propagation,” says Dr. Alexander Fandakov, who leads the R&D team working on battery electric vehicle powertrain development at IAV. “First and foremost is the safety of the vehicle, but another big reason is legislation.”
Many jurisdictions around the world have enacted legislation stipulating that if damage to the electric vehicle battery is imminent, there must be sufficient time for drivers and passengers to stop and exit the vehicle before thermal propagation occurs. For example, UN regulations require that vehicle occupants receive a signal “5 minutes prior to the presence of a hazardous situation inside the passenger compartment caused by thermal propagation”.1 Moreover, there has been discussion in some jurisdictions about significantly increasing the duration of time required between thermal runaway and thermal propagation, which would essentially mean that no thermal propagation would be allowed.
To meet these legislative requirements, manufacturers must conduct extensive testing of their battery modules or packs under different conditions to evaluate the risk of thermal runaway and devise methods to mitigate thermal propagation. Extensive testing, however, doesn’t come cheap.
“The problem with electric vehicle battery development is that when you want to perform testing related to thermal propagation, you generally need at least a module, or the entire battery pack, which typically are not available until a late stage of the development process. And when looking into thermal propagation, you have to consider different boundary conditions, and then you basically put the battery pack in the trash after the test. So these tests are very, very expensive,” Alexander explains, “and the implementation of additional propagation mitigation measures based on the test results are typically anything but straightforward at this late development stage.”
It follows naturally, then, that if you can cut down on the number of physical tests, you can save a significant amount of time and money. This is where computational fluid dynamics (CFD) comes into play. CFD allows engineers to simulate battery packs with different chemistries, materials, and configurations under different conditions to virtually assess the efficacy of thermal propagation mitigation strategies. To be an effective development tool, however, you need to have a predictive CFD code—which is why IAV elected to use CONVERGE.
“CONVERGE uses a physics-based approach to model 3D thermal runaway and thermal propagation,” says Kislaya Srivastava, Principal Engineer at Convergent Science. “This means that we don’t rely on experimental profiles, instead using chemical reaction mechanisms coupled with high-fidelity models to predict the thermal runaway behavior.”
IAV and Convergent Science worked together to develop and validate a numerical approach in CONVERGE to simulate thermal propagation, starting with modeling the thermal runaway kinetics of different battery chemistries, then using the validated kinetic mechanisms to predict the 3D spatial temperature distributions and heat transfer in battery systems employing a variety of different materials. In this blog post, we’ll take a look at an overview of the team’s 3D modeling work; for details on the experimental work and more in-depth information on the simulation studies, please refer to Sens et al. 2024.2
The team from IAV and Convergent Science first used CONVERGE to conduct single-cell tests of different lithium-ion battery chemistries, including nickel manganese cobalt (NMC) and lithium iron phosphate (LFP), as well as a sodium-ion battery (SIB). Figure 1 shows the single-cell geometry, including clamps, used for the 3D CONVERGE simulations.
The cell is modeled as a single solid with an applied anisotropic thermal conductivity along the direction of the cell layers, and interfaces are defined between the components depicted in Figure 1 to allow heat transfer between them. The team used established thermal runaway mechanisms available in CONVERGE and calibrated them to accurately represent the thermal abuse within the cell.
In this blog post, we’ll focus on an NMC811 cell, for which the team employed the Ren mechanism.3 The team calibrated the mechanism using experimental data from a constant heating test, then validated the NMC model with the calibrated mechanism against experimental data from heat-wait-seek and nail penetration tests.
Figure 2 compares the CONVERGE results with measurement data from three thermocouple positions for the constant heating and heat-wait-seek tests. Figure 3 shows the results of the nail penetration test, comparing CONVERGE with measurement data from the most representative thermocouple position. The CONVERGE results match well with the experimental data in all three cases, demonstrating that the calibrated mechanism is able to represent the thermal runaway behavior of the cell for different initiation methods, thus confirming its predictivity.
With the single-cell validation completed, the team moved on to conduct thermal propagation studies in a seven-cell module, as shown in Figure 4. They employed the same calibrated Ren mechanism for the NMC811 cell chemistry from their single-cell studies. They looked at several different scenarios, including where the space around the cells within the housing was filled with either nitrogen or air, and the application of an insulating inter-cell element or immersion oil cooling to delay thermal propagation.
“In addition to the thermal runaway chemistry, CONVERGE offers a number of features that make these simulations possible,” says Kislaya. “CONVERGE’s autonomous meshing easily handles the complex battery pack geometries with no user meshing time, and Adaptive Mesh Refinement dynamically adjusts the mesh throughout the simulation to capture the complex physical phenomena at a lower computational cost. In addition, CONVERGE’s conjugate heat modeling allows us to analyze heat transfer between the solid and fluid domains, and its multi-phase modeling capabilities enable us to investigate liquid cooling techniques.”
Figure 5 shows the results for a case with air surrounding the cells and no thermal insulation applied. Thermal runaway is initiated in the center cell (cell 7) via nail penetration; the adjacent cells also go into thermal runaway immediately after the nail penetration occurs. CONVERGE is able to capture the timing and duration of the thermal propagation very well. While the predicted peak temperatures are lower than the measured peak values, the measured peak temperatures are considered mainly as gas temperatures and thus cannot be directly compared with the solid cell surface temperatures obtained from the simulation. The solid cell surface temperatures drive the processes inside the cell that ultimately result in thermal runaway, and the simulation captures these important values.
After validating the CONVERGE model for a case with immediate thermal propagation, IAV turned their attention to mitigation strategies. In this post, we’ll focus on the adoption of an insulating inter-cell element, but the results of oil cooling can also be found in Sens et al. 2024.2
“One way you can prevent heat from transferring from one cell to another is by inserting a foam, for example, that is a thermal insulator between the cells,” says Alexander. “But such a foam also has challenges because it has an impact on the overall weight of the battery, it has an impact on cost, and so on. It’s a very complex problem, and that’s why it is crucial to be able to investigate different types of inter-cell materials with simulation.”
Figure 6 shows the impact of adding an inter-cell element on thermal propagation. Thermal runaway is once again triggered in the center cell (cell 7) via nail penetration. As you can see, the addition of the inter-cell element significantly delays thermal propagation to the adjacent cells. Overall, the CONVERGE simulations are able to represent well the progression of the thermal propagation, especially considering the immense complexity of the events occurring in the experimental setup that are not considered in these simulations, such as mechanical deformation, material melting, and material ejection out of the battery. The deviation between the measured and simulated total propagation duration is approximately 10%.
This successful collaboration brought together IAV’s extensive industry expertise and state-of-the-art testing facilities with CONVERGE’s predictive simulation capabilities. Together, IAV and Convergent Science developed and validated a numerical model to study thermal runaway and thermal propagation in battery modules. In the future, this methodology can be applied to different battery chemistries, module configurations, and thermal propagation mitigation strategies. Having a powerful and efficient method to study thermal propagation is a game-changer for industry, enabling manufacturers to meet legislative requirements and ensure the safety of electric vehicles for consumers, all while saving time and reducing development costs.
Learn more about this collaborative work in our joint webinar: A Cool Take on Hot EV Batteries: Navigating Thermal Propagation With CFD Based on Thermal Runaway Kinetics Modeling.
[1] United Nations, “UN Regulation No 100 – Uniform Provisions Concerning the Approval of Vehicles With Regard to Specific Requirements for the Electric Power Train,” E/ECE/Rev.2/Add.99/Rev.3.
[2] Sens, M., Fandakov, A., Mueller, K., von Roemer, L., Woebke, M., Tourlonias, P., Mueller, T., Burton, T., Srivastava, K., and Senecal, P.K., “From Thermal Runaway to No Thermal Propagation,” 45th International Vienna Motor Symposium, Vienna, Austria, Apr 24–26, 2024.
[3] Ren, D., Liu, X., Feng, X., Lu, L. Ouyang, M., Li, J., and He, X., “Model-Based Thermal Runaway Prediction of Lithium-Ion Batteries From Kinetics Analysis of Cell Components,”Applied Energy, 228, 633-644, 2018.
