Within Cell Integration, the focus switches almost exclusively to safety considerations in terms of the key battery attributes identified in What Matters in Battery Technology Development?, and addressing the challenge of mitigating the risk of thermal runaway.
Cells generate heat on use, and they generally produce more heat as they age. When they are packed closely together within a module or pack, heat dissipation will be reduced. This can lead to a scenario where the heat generated exceeds the heat dissipated, leading to a potential thermal runaway situation, unless additional measures are implemented. In addition, the battery can also be subject to additional stresses and operated under a range of conditions, which may accelerate the rate of self-heating. These various potential causes of thermal runaway are summarized in Figure 16, which is a more detailed version of that presented in Figure 7, and will be referred to throughout this section.
Figure 16: Detailed schematic illustrating the causes of battery failure
Therefore, there are two distinct elements to cell integration:
- Managing the heat generated as part of the normal use of the battery
- Implementing mitigation strategies to avoid the cell being subjected to extreme stress conditions
This translates into two areas of evaluation:
- Determining thermal management requirements for the battery
- Defining the safe operating limits of the battery
Although both relate to battery safety, they require two different techniques:
- Thermal management involves probing the battery response in the Optimum Zone. Therefore, an isothermal calorimeter is the instrument of choice for this analysis
- Defining the safe operating limits involves pushing the battery beyond its safe boundaries, identifying where those boundaries are, and understanding the extent of the hazard if they are exceeded. In this case, an adiabatic calorimeter should be chosen as the battery is being tested in the Dangerous Zone
This is illustrated schematically in Figure 17.
Figure 17: Safe Operating Limits can be determined from probing battery performance in the Dangerous Zone, whilst data for Thermal Management requirements can be derived from probing the Optimum Zone of operation
The third element related to addressing the challenge of mitigating the risk of thermal runaway focuses on developing a deeper understanding of the mechanism of thermal runaway and thermal propagation. This knowledge can then be used to inform management of thermal behavior and also provide insight for future cell development.
Determining Thermal Management Requirements
Cells will generally produce heat on discharge and, as they age, the heat generated will slowly increase due to the increase in internal impedance of batteries on repeated cycling. In other words, a cell’s electrical performance will decline over use. Furthermore, the cells within a module may not exhibit uniform properties upon cycling . This can cause a potential imbalance within the module, which could trigger a safety hazard and/or affect battery electrical performance and life span.
Without careful management, this self-heating can result in localized overheating and trigger a thermal runaway. It is also important that the battery is carefully managed to minimize the degradation over time and the impact this has on performance . Operating the battery at high temperatures will lead to side reactions and increase the internal rates of thermal reactions. This can lead to even greater heat generation, higher temperatures, and possibly a thermal runaway condition. These scenarios are illustrated in Figure 16.
Therefore, it is important to be able to determine thermal behavior over different operating and environmental conditions, so that the appropriate thermal management requirements can be implemented.
Typically, battery systems use convective thermal management to avoid the development of a rapid increase in cell temperature. The physical arrangement of cells within a module or pack are essential in governing heat transfer, as is optimizing the heat capacity and heat transfer rate of the coolant or heat sink used. The coolant is usually either air, liquid, or a phase-change material (PCM). The thermal management system could range from the inexpensive use of a small fan to circulate air through the battery chamber, to a more expensive option involving the utilization of a temperature-controlled liquid cooling circuit which passes across the cells in a pack . Whereas the former might not allow the battery to be used at high charging or discharging rates under all conditions, the latter could allow the temperature to be regulated evenly across the cells in the pack, allowing for the battery to be utilized under a wider range of operating conditions and temperatures. A PCM can form part of a structure that is integrated into the pack, so it is in direct thermal contact with the cells, as illustrated in Figure 18. It adds additional mass (and cost) to the system, but this extra thermal mass helps to absorb the heat given off by cells, slowing their temperature rise. Depending on the design, it can sometimes melt if the temperature continues to rise, limiting the peak temperature. The structure also provides a fast heat transfer through the battery module, minimizing the temperature difference between cells in a module .
The choice of thermal management strategy will be dependent on the thermal behavior and electrical performance of the cells and module being used, and the device’s end application. Isothermal calorimetry enables cells, modules, and packs to be characterized over a range of temperatures and operating conditions. The data generated can then be used to help select the most appropriate form of thermal management.
Figure 18: schematic illustrating a battery module with PCM/graphite thermal management, adapted from .
Defining Safe Operating Limits
Until now, we have only considered managing the thermal challenges posed by normal operation of the battery. It is also vital to understand how the battery will respond to other stress conditions which it could be subjected to during operation, so that these scenarios can be mitigated and avoided. Subjecting batteries to abuse, or stress tests is the focus of a range of battery safety testing regulations  . There is a lot of commonality between these regulations, and they cover the same kind of stress tests; for more information on these regulations, please see reference .
There are three main sources of stress that the battery or cell can be subjected to: thermal, electrical, and mechanical. A more detailed figure illustrating these potential causes of cell failure is given in Figure 16. As can be seen, with the exception of a scenario where the puncture test causes the cell to be forced open, all other causes of cell failure will result in increased self-heating, which can lead to thermal runaway, and then ultimately, to a cell fire or explosion. This underlines the importance of understanding cell thermal behavior under different stress conditions.
Thermal stress tests provide the data to help define the safe working temperature of the battery . The biggest safety concern relates to the development of a thermal runaway situation, whereby chemical reactions triggered in the cell exacerbate heat release, potentially resulting in fire or explosion. As noted previously, the cells self-heat under normal operation and within a battery are packed closely together. Therefore, an increase in temperature from external sources is likely to accelerate self-heating, potentially leading to a thermal runaway if mitigation strategies are not in place.
Taking the specific example of Li-ion type batteries, an increase in temperature will cause the rate of electrochemical reaction to increase, but it is also accompanied by a more rapid accumulation of waste products from side reactions within the battery chemistry. These side reactions will result in a faster reduction of capacity. Furthermore, as the temperature continues to increase (around 85-120 °C), the SEI (solid electrolyte interface) will start to decompose. As described below, this then triggers a sequence of potential events, the main points of which are illustrated in Figure 19 :
- The resulting increase in temperature from the SEI decomposition may then trigger electrolyte flashpoints and decomposition of the cathode and anode
- The exposed lithiated anode materials then react rapidly with the electrolyte
- The high temperature will also increase the dissolution of transition metals ions into the electrolyte, which leads to capacity degradation
- At 90 °C, the SEI will typically decompose entirely and the side reactions will increase, which will generate a lot of heat and gas. Most functionality will be lost at this point.
- As the temperature increases further, the melting point of the separator is reached, and all functionality will be lost
- The electrolyte then decomposes and produces gas. Reactions between the active materials of the anode and the binder may follow
- The cell may then burst from the increased pressure generated, and toxic gases and electrolyte solvents can leak out
The acceleration in temperature ultimately leads to a thermal runaway of the battery system. Therefore, efficient thermal control of the battery is critical. If the heat generated can be safely dissipated, the series of exothermic processes can be avoided, and a thermal runaway situation averted.
Figure 19: typical cascade of potential effects when a Li-ion cell overheats
Although the focus is often on high temperatures and overheating, low temperatures can also induce a thermal runway, in extreme cases.
A number of challenges result from operating batteries at low/freezing temperatures :
- Slower kinetics, leading to a slower rate charge transfer
- Lower electrolyte conductivity
- Increased SEI resistance
- Low solid-state lithium diffusivity in graphite
This will lead to charging difficulties, capacity fade, and power attenuation. It can also lead to lithium plating at the anode (graphite surface) during charging. These lithium deposits are no longer involved in subsequent electrode reactions, which is the main reason for the capacity fade. The deposits can also form dendrites, which can pierce the separator and induce an Internal Short Circuit (ISC) and subsequent thermal runaway.
Adiabatic calorimeters are ideally suited for performing thermal stress tests. They incrementally heat the cell or battery until self-heating is detected. They then maintain adiabatic conditions so that the cell or battery continues to self-heat, possibly until destruction. The technique allows the onset temperature of the thermal runaway to be detected and the rate of heat generation during cell failure to be determined accurately due to the well-defined and controlled test conditions. This information can be used to help define the safe working temperature of the battery. In addition, in-situ thermal mapping can be used to highlight regions of the battery which generate more heat when subjected to external thermal stresses . Case Study 3 describes an investigation using the BTC-500 to determine the onset temperature of a Li-ion pouch cell.
Electrical stress tests, such as over-charging and over-discharging, can help facilitate the determination of the maximum or minimum safe voltage and maximum safe current of a cell or battery.
Over-charging is usually defined as when a battery is forcibly charged beyond its cut-off voltage, and is reportedly one of the main causes of EV battery failure . It causes over-delithiation of the cathode, which leads to the breakdown and collapse of the material’s structure. This then leads to side reactions between the cathode and the electrolyte – it is this that then causes the rapid increase in temperature in the cell. Often the reactions can be very violent, releasing a lot of gas and heat .
At the same time, over-charging will result in more rapid dissolution of the transition metal from the cathode into the electrolyte, which will then deposit on the anode surface. This plating will cause a significant increase in impedance, resulting in capacity fade of the anode  .
Similarly, the anode will not be able to accept the greater level of lithium ions now available due to the over-charge. This will result in metallic lithium depositing on the anode surface, which will then react with the electrolyte. This leads to a further impedance increase and heat generation . The deposited lithium can also form dendrites, which can puncture the separator and cause an internal short circuit (ISC) and result in a thermal runaway situation, as depicted in Figure 16. Figure 20 illustrates the consequences of over-charging an 8 Ah Li-ion pouch battery based on NMC-graphite chemistry to 5V (instead of the safe maximum voltage of 4.5V).
Adiabatic calorimeters can be fully integrated with charge-discharge units to conduct overcharging electric stress tests. This not only enables the maximum safe voltage to be determined, but it also provides insight into the severity of the consequences when these safe operating limits are exceeded. This is because the instruments allow the test to be conducted under adiabatic conditions, simulating a worst-case scenario when no dissipation of the heat generated by the battery has taken place. This allows the consequences of malfunction to be realistically measured and as the test environment is well-defined, it allows comparison between different batteries.
In order to increase robustness to over-charging, research efforts within Cell Development are focused on improving the structural stability of the cathode and the thermal stability of the electrolyte at high voltage. For example, lithium iron phosphate (LFP) is much more thermally stable than the likes of NMC or LMO (however, it does have a lower energy density, again demonstrating the trade-off between battery attributes).
Figure 20: Li-ion NMC-graphite battery after over-charging to 5V in an adiabatic BTC
Exceeding the maximum safe current of a battery (i.e., over-discharging) could lead to the cell overheating and potentially a thermal runaway situation. Discharging a battery to a voltage below its lower limit will lead to the decomposition of the SEI. In doing so, it will lead to cell swelling as gases are generated. It will also indirectly lead to a reduction in capacity, as subsequent recharging steps will lead to the formation of new SEIs, which will consume active lithium ions and electrolyte. Similarly, an over-discharge will cause the copper current collector of the anode to be oxidized. It will subsequently dissolve in the electrolyte and the copper ions will migrate to the cathode before forming copper deposits on the cathode during battery charging. The deposits can form dendrites on the cathode, which can cause the separator to be penetrated, inducing an internal short circuit (ISC) and potentially a thermal runaway.
Therefore, it is vital to be able to determine what the maximum safe current of a battery, or C-rate, is so that a suitable overcurrent protection device can be implemented to limit the current. Defining what the C-rate is for a battery can be determined by exploring the impact of different discharge rates on a battery.
As discussed previously, adiabatic calorimeters can be integrated with a charge-discharge cycler, which allows the battery temperature during charging and discharging cycles to be measured. The system can be set up to automatically run repeated cycles, enabling information on the longer-term stability of the battery to be determined. It can be used to explore battery over-discharge behavior over a range of discharge rates . The advantage of placing the battery inside an adiabatic calorimeter like the BTC-130 and the BTC-500 during the test is that the temperature changes will accurately indicate the energy changes taking place in a worst-case scenario where no heat dissipation is taking place. The test is similar to the over-charging test, except whereas in the over-charging test the current is at a normal and safe value and the battery is charged to higher voltages, in the over-discharge test, the voltage is kept within the safe maximum voltage, and the discharging current is pushed beyond the safe operating limits. Case Study 4 describes an investigation exploring the impact of high discharge rates on Li-ion batteries using the BTC-500.
Thermal runaways can arise from both internal and external stimuli. An external short circuit will lead to self-heating and the resultant cell temperature increase can result in a thermal runaway. The BTC-130 and the BTC-500 can be equipped with the means to deliver an electrically induced external short circuit so that the impact and consequences can be evaluated. A visual record of the event can also be captured from integrated camera footage.
This is normally divided into two types of stress: crush/collision and penetration . The crush test simulates the deformation of the battery caused by, for example, a vehicle crash or by sitting down forcefully on a phone. As Figure 16 illustrates, the impact can cause an internal short circuit directly, or it can result in the generation of particles which will then induce an internal short circuit. Similarly, penetration tests, such as a nail/spike test, can cause a separator failure which then results in an internal short circuit. An internal short circuit will lead to increased self-heating and may result in a thermal runaway situation. The spike test can also cause the cell to be forced open, leading directly to material and gas ejection, possibly followed by cell fire. These tests can help provide an indication of the structural stability of the cell.
The BTC-130 and the BTC-500 can be equipped to perform a range of puncture tests within a safe environment for operators to carry out the test. Thermal data on the severity of the incident can be acquired and the integrated camera enables the unfolding event to be visually captured. Images depicting the moments immediately before and after a nail penetration test on a Li-ion pouch cell are illustrated in Figure 21.
Figure 21: immediately before and after a nail penetration test on a Li-ion pouch cell
Understanding Thermal Runaway and Thermal Propagation
Developing a deeper understanding of the mechanisms behind thermal runaway and thermal propagation can inform safety protocols in module and pack design, and also provide insight for future cell development.
In general, most extreme conditions result in thermal stress on the cell or battery, which can lead to a thermal runaway. The extent of the thermal runaway will depend on factors such as the capacity of the cell, the cell type, the cell history, the cathode/anode material, the electrode composition, and the state of charge (SOC) of the battery . Therefore, the data obtained from the stress tests performed in the BTC-130 and BTC-500 can be used to model a cell’s predicted behavior . Successive onset temperatures of decomposition of components within the cell can be detected, and the rate of heat generation during the failure determined. This data can be combined with data from structural analysis tools, such as X-ray CT imaging and scanning electron microscopy, to help decipher the series of events that a Li-ion battery undergoes during failure . The external analysis of the composition of any evolved gases collected can provide insight into the chemical reactions that are taking place during the thermal decomposition, furthering mechanistic understanding of the thermal runaway.
The BTC-500 also enables the triggering of a thermal runaway in a cell at a specific position within the module via a mechanically- or electrically-induced short circuit. This can provide information on how a thermal runaway propagates within a module, and the magnitude of the thermal event can be characterized. This information can be used to develop models of module behavior and can also be used to inform mitigation measures within the module design to ensure heat dissipation is greater than heat generation.