2.1 Abuse testing
The International Electrotechnical Commission (IEC), Underwriters Laboratories (UL) and the Japan Battery Association (JSBA) initially defined the abuse test for consumer electronics cells, simulating the extreme conditions that cells may encounter when working, usually divided into thermal abuse, Electrical and mechanical abuse. The common thermal abuse is the hot box test, the electrical abuse includes overcharge and external short-circuit experiments, and the mechanical abuse includes acupuncture, extrusion, shock and vibration. Corporate and industry standards generally describe a battery's response to abuse testing as no change, leaking, burning, exploding, etc. Temperature, gas, and voltage responses to abuse can also be recorded based on additional sensors and detection systems. The standard for the battery to pass the abuse test is that it does not burn or explode. Since the abuse test is aimed at commercial finished cells and close to real use conditions, it is currently more of a safety test standard for the battery industry than a research method.

2.2 EV-ARC test
The early ARC was only suitable for studying the thermal runaway behavior of a small number of material samples. Feng et al. developed a method to use EV-ARC to study the adiabatic thermal runaway behavior of large-volume cells. The principle and conclusion of the research method are shown in Figure 6. The heating chamber of the ARC is larger, so more precise temperature control technology and stricter calibration scheme are required. Based on the EV-ARC test, the characteristic temperatures T1, T2 and T3 of the thermal runaway of the cell can be quantitatively calibrated, corresponding to the starting temperature of the self-heating of the cell, the starting temperature of the thermal runaway of the cell and the maximum temperature of the cell, respectively. Safety provides a more accurate and quantitative evaluation index. Standardized test conditions can help to establish a unified and reliable cell thermal runaway behavior database, and analyze the thermal runaway mechanism of cells in different systems. These problems are difficult to quantitatively verify in conventional abuse tests.

Compared with ordinary heating abuse experiments, the temperature of the EV-ARC experimental environment is precisely controlled by the program, and the obtained test results have better repeatability and higher data interpretability. In recent years, it has become an important factor in evaluating and researching cell safety. means. However, the adiabatic thermal runaway environment simulated by EV-ARC is still different from the real battery abuse condition. To evaluate the actual safety of the battery cell, a large number of test methods simulating real severe conditions are still needed.

2.3 High-speed imaging technology
In order to more intuitively understand the evolution of materials and structures inside the battery during thermal runaway, researchers have developed a transmission X-ray microscopy (TXM) method that combines infrared thermometry and in situ acupuncture with auxiliary functions such as: 7(a) to (c). Due to thermal runaway, violent reactions often occur in a very short period of time, accompanied by violent phase and structural changes. This feature imposes a fairly high temporal resolution requirement on the TXM characterization method. The number of X-ray photoelectrons that can be emitted by laboratory X light sources is limited, and it takes a long time to collect a set of TXM image data.

Since the transmission projection map can only reflect two-dimensional information in a certain direction, if the distribution of matter in a real three-dimensional space is to be accurately quantified, computer imaging technology (computing tomography, CT) is required. Based on every 500 TXM reconstructions, one X-ray CT result can reach 2.5 frames per second, realizing the imaging of the internal spatial distribution of the battery with a certain time resolution. The CT results can clearly see the changes of battery materials at various stages in the thermal runaway process, such as damage to the electrode active material layer, melting and re-agglomeration of the copper current collector, etc.

Combined with the projection images obtained by TXM technology and the results of high-speed X-ray CT, it is possible to clearly understand the failure behaviors such as reaction, gas production, and structural damage of various materials at different positions inside the battery during the thermal runaway process. On the other hand, in situ experiments such as acupuncture, infrared heating, extrusion, and stretching can help to study and understand various macroscopic failure behaviors of batteries.

3 System thermal safety research
The safety of the battery system is the most direct problem facing the application of lithium batteries at present. At present, thermal runaway of commercial cells cannot be completely avoided. Preventing thermal runaway expansion at the system level is a possible safety solution. The cost of carrying out experimental research at the system level is high, but it is unavoidable. With the aid of simulation, the system design can be predicted and optimized in advance to reduce the experimental cost.

3.1 Thermal runaway expansion and fire hazard testing
The experimental research cost and risk of thermal expansion of the battery system are relatively high. The main methods are to induce thermal runaway of the battery cell through heating, overcharge, and acupuncture, and use contact thermocouples, infrared temperature measurement and other methods to study the temperature in the battery. The distribution and changes in the system, this method can only obtain local multi-point thermal runaway information. The above tests can evaluate the safety and loss of control risk of large battery packs at a practical level, providing important information for safety improvement, early warning, fire protection and disaster disposal.

3.2 Disaster gas research and early warning program design
In the process of actual use and safety failure of batteries, the composition and generation of gases are important research topics, which are closely related to the early warning of battery thermal runaway, explosion, and fire spread. In terms of material nature, the gasification of the organic electrolyte in the battery and the high-temperature side reactions of the active components will release gas. The mixed gas generated under heating conditions can be analyzed by gas chromatography-mass spectrometry spectrum, GC-MS), Fourier transform infrared spectroscopy (fourier transform infrared spectroscopy, FT-IR) and other means to analyze the composition. At present, these gas detection technologies are relatively mature, but in the process of safety research, the collection and quantification of gases still require the assistance of special containers or samplers.

4. Safety research of next-generation lithium batteries
Prevention, early warning, and prediction of battery safety depend on a deep understanding of the structure-activity relationship from system to cell to material thermal runaway. Looking at the lithium battery fire incidents that have attracted widespread attention in recent years, most of them occurred in the initial application stage of new technologies and new materials. After extensive attention, the research on the safety of this battery system has increased. The hysteresis between battery safety research and battery electrochemical performance research is a distinctive feature of battery safety research.

In order to meet the requirements of high safety and high energy density brought by the electrification wave, it is expected that non-flammable electrolytes or solid electrolytes will be used in lithium-ion batteries to completely solve the safety problems of batteries and achieve high energy density. However, battery safety is not only related to the thermal stability of the materials inside the battery itself, but also to the interaction between materials and the complex environment inside the battery.

In summary, in order to ensure the safety of batteries while developing high-energy density batteries, researchers need to simultaneously carry out prospective battery safety verification and research as soon as possible while optimizing the electrochemical performance of cells. Only by clearly and comprehensively understanding the thermal failure mechanism of the battery and the factors affecting the safety of various dimensions can the effective safety prevention of the battery be done in the application stage. Figure 8 shows the technological maturity cycle of new materials and technologies in the battery field from basic research to mass production. It can be seen that the large-scale application of a new technology requires a huge investment of manpower and material resources, and it takes decades to achieve mass production. However, the safety verification of the battery is often carried out when the battery is close to mass production, and it is often aimed at passing the battery safety test standard, and it is impossible to systematically and deeply understand the safety behavior and The internal mechanism lays hidden dangers for future safety accidents. For the early battery system, due to the low energy density, the safety problem is not prominent, and the energy density of the latest lithium-ion battery cell can reach more than 300 W·h/kg, the new technology and The new system has higher energy density. These new technologies and systems with high energy density are facing more severe safety challenges. Therefore, the safety research and verification steps of the battery should be carried out as early as possible, and the battery should be carried out as soon as possible after the basic determination of the cell structure. Safety testing and mechanism research work is expected to be ready in the early stage of the real mass production stage, to find out its safety characteristics and behavior, and to design corresponding protection and early warning measures.

At present, the material system of the next-generation chemical energy storage battery has not been finalized. The new materials that may be used in the new-generation lithium-ion battery include lithium-rich materials, lithium-free high-capacity cathode materials, silicon-based anode materials, lithium metal anode materials, and solid electrolytes. Etc., if the use of lithium metal negative electrodes is considered, the epitaxy of the lithium battery concept can be further expanded. However, from the perspective of academic reports, there are few reports on the thermal behavior of new materials and the practical safety of new systems. At present, the safety awareness of most new lithium battery systems is still in the unknown or early stage. The research methods reviewed in this paper can be used not only to study the safety of existing commercial lithium-ion batteries, but also to understand the thermal stability of new lithium battery material systems in advance from the material level, and to predict their cells and systems based on simulation methods. It has important guiding significance for selecting the technical route of next-generation lithium batteries and ensuring the smooth implementation of new technologies of high energy density lithium batteries. "