Safety of power lithium batteries for new energy vehicles
Recent electric vehicle accidents have drawn significant attention, so today I'll focus on the safety of electric vehicles. I'd like to present this from four aspects, starting with electric vehicle accident statistics. This is a summary of the causes of spontaneous combustion and fires of electric vehicles abroad in recent years, with collisions being the most prominent cause. Actually, gasoline-powered vehicles can also catch fire after a collision; this is the domestic statistics for such fires. Domestic fires have several key characteristics: First, they primarily use ternary lithium batteries, with some lithium iron phosphate batteries, but ternary lithium batteries are the most prevalent, accounting for over half. Second, cylindrical batteries are the most common type, as they have steel casings and are tightly rolled, making them prone to explosion in the event of thermal runaway, which can then ignite other batteries. Third, charging-related fires account for a large proportion of accidents. Generally, batteries won't thermally run away after discharging to a certain depth; thermal runaway usually occurs when the battery is fully charged, making it more likely to happen during charging. This is because the battery is connected to the charging system during charging, which is the most vulnerable time for thermal runaway. Short circuits in high-voltage electrical components can also contribute to accidents. Furthermore, from the perspective of vehicle models, both new and old models exist, and the battery system energy density is not very high. This is because the main cause of accidents is vehicles installed in the previous few years. Overall, the system energy density is not very high, and it is not the high energy density battery we usually think of.
Battery thermal runaway is arguably the main cause of these accidents. What is battery thermal runaway? When the battery temperature reaches a certain level, the battery will have a chain of negative reactions, an exothermic reaction, so the temperature rises rapidly. The highest rate can reach nearly 1,000 degrees Celsius per second, so it is very fast.
What causes thermal runaway? First, the battery overheats. As mentioned earlier, thermal runaway occurs when the battery gets too hot. There are various reasons for overheating, such as uneven temperature distribution within the battery pack itself, resulting in localized high temperatures, overcharging or over-discharging, external short circuits, internal short circuits, and other electrical causes that generate heat. There are also mechanical causes, such as water ingress, poor sealing, or impacts.
Let's examine the main causes of these recent accidents, which we believe are product quality issues. Product quality issues refer to the failure to strictly adhere to relevant technical standards and specifications during the design, manufacturing, verification, and use of products. There are three main categories: first, insufficient battery product testing and verification; second, changes in vehicle reliability during use; and third, problems with charging safety management technology. We will now analyze these aspects.
First, there is insufficient testing and verification of battery products. Because the subsidy reduction policy is implemented annually, it doesn't quite match the overall product development cycle. For example, improvements to our chemical material system typically take more than a year, but because the company blindly pursues high specific energy by following subsidy policies, it has shortened the testing and verification time. Sometimes, to shorten the development cycle, physical improvement methods are preferred, such as thickening the battery active material and thinning the separator. This increases the battery's specific energy, but reduces safety performance.
Secondly, the methods for testing and verifying electric batteries are not perfect and cannot reflect the actual usage conditions of vehicles. A large number of companies have not established internal battery safety testing standards, and some companies do not even have the ability to conduct battery safety testing, resulting in inconsistent quality of the products they produce.
The third reason, as mentioned earlier, is the decrease in reliability during vehicle aging. For example, the waterproofing effect deteriorates throughout the entire lifespan. While batteries are typically designed to meet IP67 standards, the seal weakens over time, allowing water to enter and potentially causing short circuits. Another example is the laser-welded connectors in the battery; gaps can easily appear inside the weld points, increasing impedance and leading to overheating and thermal runaway. Furthermore, the aging of the battery system and charger's high-voltage electrical components, such as the contactors during charging (which may frequently open and close, sometimes arcing), can cause overheating, burns, or adhesion on the contactor surface, resulting in short circuits and overheating – all contributing factors to thermal runaway.
The fourth reason is charging itself. Data communication during charging is not standardized, and BMS and charger manufacturers have not strictly implemented the newly promulgated national standards. Regarding functional safety, our battery management system should theoretically have excellent power-off functions, controlling the charging process up to the desired point. However, we are not strictly adhering to functional safety standards, specifically ISO 26262. This is another reason for our failure to comply. We also fail to strictly implement relevant charging safety standards. For example, charging relays should have diagnostic functions for adhesion, but some lack these to save costs. The battery management system and charging station are not equipped with qualified insulation testing devices. The charging circuit formed by the vehicle and charging station does not meet the standard requirements for insulation voltage, creepage distance, overload, IP rating, insertion and extraction force, locking, temperature rise, lightning strike, and other indicators. The BMS has not strictly followed charging guidance specifications. Why is this a quality issue? It's because we haven't strictly adhered to standards and specifications in the design, manufacturing, use, and verification stages. Of course, we also lack some aspects, such as annual safety inspections, which are missing, but this is not just a company issue; it's a government responsibility.
High-energy-density batteries face more severe safety technology challenges, so I will talk about this aspect below.
Based on the development trend of specific energy of lithium batteries for new energy vehicles in my country, we will soon move towards high-specific-energy batteries with a capacity of 300 Wh/kg. These products, namely the so-called high-nickel ternary 811 batteries, will soon enter the market. These high-specific-energy batteries will face higher safety requirements than the previous relatively low-specific-energy batteries. In this regard, Tsinghua University has established a dedicated battery safety laboratory to conduct related basic research and technological development. Here, we will briefly introduce the research results for your reference.
Currently, Tsinghua University's Battery Safety Laboratory has established extensive collaborations with domestic and international companies and research institutions, including major companies such as BMW, Mercedes-Benz, and Nissan.
The research focuses on three aspects of thermal runaway: first, the causes of thermal runaway, including thermal, electrical, and mechanical factors; second, the underlying mechanisms of thermal runaway, enabling protection through material design; and third, thermal propagation. If a single cell cannot prevent thermal runaway, secondary protection measures are necessary, namely, cutting off the propagation of thermal runaway at the system level. Cutting off propagation can prevent accidents. Our control of thermal runaway in high-energy-density batteries relies not only on the materials themselves but also on system-level measures.
The first focus is on the mechanism and suppression of thermal runaway. We conducted our research using two experimental methods: differential scanning calorimetry (DSC), used for studying the thermal stability of materials, and accelerated calorimetry (ECC), used for measuring the thermal runaway of individual battery cells.
Several characteristic temperatures of thermal runaway in high-energy-density batteries. Generally, when the battery temperature rises to a certain level, the battery will generate heat itself; we call this temperature T1. When heat generation becomes uncontrollable, thermal runaway is triggered, called T2. Finally, when the temperature rises to its highest point, we call it T3. The important point where the thermal runaway mechanism is unclear is the stage between T2 and T3. It is generally believed to be caused by an internal short circuit, which is indeed the case for conventional batteries. However, our research found that this is not entirely true for high-energy-density batteries. We found that thermal runaway still occurs even without an internal short circuit. This is because the high-temperature resistant new separator in high-energy-density batteries does not change above 200 degrees Celsius, and the electrolyte is almost completely evaporated. However, at 230-250 degrees Celsius, the oxygen released during the phase transition of the positive electrode material reacts with the negative electrode, resulting in an exothermic peak.
Let's also examine the differences between ternary lithium-ion batteries with varying nickel contents. Compared to the commonly used 622 or 532 batteries, the exothermic peak of the 811 battery is significantly higher, indicating that the 811 has poorer thermal stability. Our preliminary conclusion from this analysis is that high-nickel cathodes have a significant impact on overall battery safety, while silicon-carbon anodes have little impact on safety initially, but a greater impact after cycle degradation.
There are a series of ways to address this poor thermal stability, such as material coating. We have also discovered a new method, which is to use monocrystalline particles to replace polycrystalline cathode materials. This has greatly improved the thermal stability of the battery and correspondingly improved its safety.
The second is thermal propagation. The real accident is caused by thermal propagation, which means that after one battery cell thermally runs away, the entire battery pack spreads the heat, and a fire occurs.
Based on our heat transfer analysis through testing and simulation of the thermal runaway propagation process, we designed a thermal insulation method: adding thermal insulation material along the dominant heat transfer path. Experiments have shown that this method effectively prevents the spread of thermal runaway. This firewall technology has been adopted in international regulations on thermal runaway propagation in electric vehicles, which are advocated by my country.
The third aspect concerns the causes of thermal runaway and battery management. The first cause is internal short circuits. Analysis of in-use and accident-damaged batteries revealed that after a period of use, the uniform electrode sheets developed during battery manufacturing can crack in folded areas, easily leading to localized lithium plating and thus thermal runaway. Additionally, impurities during manufacturing can also cause internal short circuits; we call this the "cancer" of the battery because we don't know when it will trigger thermal runaway, and sometimes it occurs after a long period. To address this, we invented an alternative experimental method for internal short circuits in batteries, achieving the desired internal short circuit by implanting a shape memory alloy into a specific battery. Our research categorized internal short circuits into four types, with the connection between the aluminum current collector and the negative electrode being the most dangerous. This type requires early warning, and we conducted a series of studies, obtaining a three-stage evolution process for internal short circuits. In the first stage, only the voltage decreases without a temperature increase; in the second stage, a temperature increase occurs; and in the third stage, a rapid temperature rise occurs, which is thermal runaway. Based on this evolution process, we strive to identify internal short circuits in the first two stages, which can provide an early warning of internal short circuits that may lead to thermal runaway 15 minutes in advance. This technology has already been developed in cooperation with CATL.
The second aspect is charging. Through testing and analysis, we clarified the overcharge thermal runaway mechanism. Based on this, we used a thermoelectric coupling model to predict the performance of battery overcharge thermal runaway. Overcharge accidents are generally micro-overcharges, such as those caused by battery inconsistencies. Because of these inconsistencies, some areas are fully charged while others are not, leading to some fully charged batteries being slightly overcharged. This results in lithium dendrite formation on the negative electrode material, which is known as lithium plating. This deteriorates safety and can cause short circuits.
To address this issue, we developed a lithium-free fast charging technology based on a reference electrode. This technology controls the potential of the negative electrode above zero (lithium deposition occurs below zero), requiring the addition of an extra electrode, resulting in a three-electrode system. Based on this three-electrode system, feedback and observation can be performed using a model. This is our lithium-free fast charging technology. After applying this technology, lithium deposition is eliminated, and the charging speed is significantly increased.
The third reason is aging. As batteries age, inconsistencies increase, which is why the inconsistencies become more pronounced with each new battery cycle. As capacity consistency deteriorates, the accuracy of battery management also suffers. Furthermore, aging at low temperatures severely impacts battery thermal stability, lowering the autogenous heating temperature at which thermal runaway occurs, making thermal runaway more likely.
Through analysis of these issues, we found that the core of ensuring battery system safety lies in the development of advanced battery management systems. Currently, domestically produced battery management systems lack sufficient functionality and accuracy, especially in terms of safety features. Therefore, it is necessary to increase R&D efforts in battery management systems. Tsinghua University has accumulated rich experience in battery management systems, having already obtained 65 patents. These patents have been applied in collaborations with renowned domestic and international companies, and some have even been licensed to Mercedes-Benz.
So how do we completely solve the battery safety problem? In the short term, we can ensure safety through some technologies, but in the long run, ensuring absolute battery safety requires forward-looking scientific research. High energy density in lithium-ion power batteries is a global development direction and trend. We cannot stop developing high-energy-density batteries because of safety issues; the key is to find a balance between high energy density and safety. For example, the intrinsic safety problem of high-nickel ternary lithium-ion power batteries is that the positive electrode releases oxygen. We can delay oxygen release from the positive electrode and improve stability by modifying the interface; another approach is to develop next-generation solid-state electrolytes to fundamentally solve the problem of electrolyte combustion.
Based on a comparison of lithium-ion battery technology routes in various countries, lithium-ion batteries using liquid electrolytes will continue to evolve towards solid-state batteries in the short term. Considering both battery cost and the future development direction of lithium-ion batteries, we suggest that my country should follow a similar path: using liquid electrolytes in the short term, developing high-nickel ternary cathodes and silicon-carbon anodes, and preventing safety accidents through battery management systems and thermal runaway suppression. These batteries could meet the 500km range requirement for electric vehicles. In the medium to long term, the transition from liquid electrolytes to all-solid-state batteries is expected, with industrial-scale applications anticipated by 2030.
In conclusion, we must strive to resolve the intrinsic safety issues of power lithium batteries to ensure the healthy development of the new energy vehicle industry. My report can be summarized as follows:
We should view the recent fires involving new energy vehicles in a proper light. The main reasons for these incidents are product quality issues, failure to comply with technical specifications and standards, and excessively short technical verification cycles.
Policy recommendations include:
First, the original industrialization target (350 Wh/kg for individual units, 260 Wh/kg for systems, and 2000 cycles by 2020) is too high. From a safety perspective, I think it is not advisable to force it.
Second, subsidy policies should conform to the laws of technological development. The increase in energy density should not be too rapid or changed too frequently. This is my suggestion to the Ministry of Finance.
Third, we should expedite the introduction of annual safety inspection standards for electric vehicles. Additionally, to better handle and analyze electric vehicle accidents, it would be ideal to have a black box for the vehicle, and the battery pack should have fire safety interfaces. Currently, the battery packs are tightly sealed, making firefighting extremely difficult. These are suggestions to the Ministry of Public Security.
Finally, I believe battery safety is the primary focus of revolutionary breakthroughs in battery technology and the key to improving the performance of pure electric vehicles. The further the battery industry develops, the more battery safety becomes a bottleneck technology. For example, fast charging technology that can charge more than 300 kilometers in ten minutes poses a challenge to battery safety. Increasing the voltage from 300V to 600V or even 800V is also related to safety and will be the main battleground for competition in the future pure electric vehicle market. It can be said that safety is the lifeline for the sustainable development of electric vehicles. National scientific and technological research and development of power lithium batteries must focus on safety, comprehensively improve the safety technology of existing lithium-ion power lithium battery systems, and strive to achieve breakthroughs in new solid-state battery technology.
