Experimental Study on Thermal Problems of Lithium-ion Batteries

The thermal characteristics of lithium-ion batteries directly affect their application performance (capacity, internal resistance, power, etc.) and thermal safety, which is a core concern for consumers. To guide battery design and usage strategies and ensure their safe and efficient application, in-depth research on thermal characteristics under different operating conditions is crucial. This paper comprehensively summarizes and analyzes the research progress on lithium-ion battery thermal problems from both experimental and model simulation perspectives, pointing out the advantages and disadvantages of the two methods and proposing suggestions for future research combining both approaches.

Currently, commonly used rechargeable batteries include lead-acid batteries, nickel-cadmium batteries, nickel-metal hydride batteries, and lithium-ion batteries. Among them, lithium-ion batteries are widely used in consumer electronics, power batteries, and energy storage due to their advantages such as long cycle life, high charge-discharge efficiency, high specific energy, and no pollution. However, in recent years, frequent safety accidents such as fires and explosions of lithium-ion batteries have made thermal safety risks a bottleneck for their further development. Overcharging and over-discharging of lithium-ion batteries can easily cause dendrites to penetrate the separator, leading to short circuits, or cause internal short circuits due to compression or puncture, both of which result in a large accumulation of heat, a rapid rise in temperature, and ultimately thermal runaway. Therefore, studying battery thermal characteristics and thermal safety, optimizing battery design, estimating internal temperature changes, and developing thermal management schemes are of great significance for ensuring the safe and reliable operation of batteries, extending their service life, and avoiding thermal runaway accidents. Currently, research on lithium-ion battery thermal issues is mainly divided into two categories: experimental research and model simulation.

1. Experimental Research on Thermal Generation in Lithium-ion Batteries

Experimental methods are the core means of studying the thermal generation of lithium-ion batteries. They mainly utilize calorimetric equipment to monitor the thermal characteristics of the battery under specific operating conditions, accurately obtaining thermal generation data to provide fundamental support for subsequent research.

1.1 Experimental Research Using Combined Calorimetric Equipment

Currently, the core equipment for lithium-ion battery thermal generation experiments is the accelerated calorimeter (ARC) and the isothermal calorimeter (IBC). The ARC is used to test the exothermic behavior and safety of batteries and components under near-adiabatic conditions, and can conduct tests such as thermal stability, material thermal properties, specific heat capacity, thermal runaway visualization, and needle penetration/squeeze/overcharge tests. The IBC maintains a constant battery temperature through a cooling system, accurately measuring the heat exchange between the battery and the external environment under normal operating conditions and within a typical temperature range. Current research often combines calorimetry with electrochemical testing methods to explore the intrinsic relationship between heat generation and electrochemical behavior.

Using 18650 cylindrical lithium-ion batteries as the research object, a calorimeter and a multi-channel battery cycler were used to analyze the effects of operating temperature (35℃, 45℃, 55℃) and charge/discharge rate (C/3, C/2, C/1) on the heat generation rate. The results showed that the battery continuously releases heat during discharge, and initially absorbs heat followed by release during charging (initial reaction heat dominates, later Joule heat dominates). Furthermore, the discharge rate has a significant impact on the exothermic effect, while ambient temperature has a minor effect. Expanding the battery types, three different manufacturers' 18650 cylindrical batteries were selected to investigate the effects of charge/discharge rate on temperature rise and heat generation rate at 35℃, verifying the significant influence of the discharge rate, consistent with previous research.

Using a 20 A∙h lithium iron phosphate square battery as the research object, an isothermal/adiabatic calorimeter and a charge-discharge tester were used to systematically analyze the effects of charge-discharge rate (0.5C~2C), ambient temperature (-10℃~40℃), and state of charge (0~70%) on thermal characteristics. The results show that under isothermal conditions, the higher the charge-discharge rate, the smaller the state of charge, and the lower the ambient temperature, the higher the battery's heat generation power and temperature change rate. Under adiabatic conditions, the higher the charge-discharge rate, the more significant the temperature rise. The state of charge only affects the temperature change rate during the discharge phase; the higher the initial temperature, the lower the temperature rise. This provides data support for selecting battery operating conditions.

1.2 Theoretical Calculation to Aid Experimental Analysis

The theoretical calculation method is based on the principle of heat generation. By measuring key parameters such as overpotential, entropy coefficient, and internal resistance, and combining them with formulas, the total heat generation of the battery is estimated. During normal charge-discharge, the heat from side reactions and mixing processes can be ignored. The heat generation rate can be calculated using the Bernardi simplified model. The core requirement is to determine the battery's internal resistance (Rin) and entropy coefficient (dU/dT). The internal resistance of a battery is affected by temperature, state of charge, and aging, with clear patterns, but variations exist due to differences in battery materials and manufacturing processes.

Two 18650 cylindrical batteries were selected, and their resistance under different states of charge was tested using four methods. The V-I characteristic curve method yielded results consistent with and higher than the open-circuit voltage-operating voltage difference method. Simultaneously, entropy change was tested using both methods, showing high data agreement. The estimated temperature rise and heat generation rate, combined with the resistance and entropy change data, largely matched the experimental results, verifying the feasibility of the calculation method.

2. Development of Lithium-ion Battery Thermal Models

With the development of computer technology, model simulation has become an important tool for studying the thermal problems of lithium-ion batteries. Based on dimensionality, models can be categorized into lumped-mass models and one- to three-dimensional models; based on mechanism, they can be categorized into electrochemical-thermal coupling models, electrothermal coupling models, and thermal abuse models. Each model addresses thermal problems in different scenarios.

2.1 Electrochemical-Thermal Coupling Model

This model is constructed from the perspective of heat generation from electrochemical reactions and is suitable for simulating the temperature distribution under normal battery operating conditions. It typically assumes uniform current density (reliable accuracy for small batteries, but errors exist for large batteries). A pseudo-two-dimensional electrochemical model coupled with a three-dimensional heat transfer model, considering heat sources such as electrochemical reactions, polarization processes, and ohmic losses, yielded simulation results for a 10 A∙h lithium iron phosphate pouch battery that were consistent with experimental and infrared testing results, validating the model's effectiveness. It was also found that the battery temperature exceeded 50℃ during 5C discharge, necessitating the design of cooling measures.

A one-dimensional electrochemical and three-dimensional thermal coupled model was established to study the thermal behavior of LiMn2O4 batteries. It was found that reversible heat is not negligible at low discharge rates, while ohmic heat dominates at high discharge rates. Reducing electrode thickness and active material particle size can lower the battery temperature. For 18650 cylindrical batteries, a cylindrical coordinate heat generation model was used to explore the thermal characteristics at different discharge rates. Simulation and experimental results showed good agreement, confirming that Joule heating dominates at high discharge rates and entropy change heating dominates at low discharge rates.

2.2 Electrothermal Coupled Model

This model combines the internal current density distribution of the battery to study the temperature field distribution, guiding the design and consistency research of battery shape, electrodes, and current collectors. Currently, most models use two-dimensional or three-dimensional non-layered models, and there is still room for improvement in accuracy. A two-dimensional electrothermal coupling model was used to study LiMn2O4 and Li[NiCoMn]O2 polymer batteries, respectively. The effects of electrode structure and discharge/charge rate on potential, current density, and heat generation rate were analyzed. The simulation results showed good agreement with experimental data, providing support for the optimization of cooling strategies.

For a 14.6 A∙h LiMn2O4/C battery, an electrothermal coupling model was established to analyze low-temperature discharge behavior. By modifying the model parameters, the simulation results at low temperatures (-20℃~0℃) were made consistent with the experimental results. Constant power charge-discharge simulations were conducted to obtain the temperature distribution under different power levels, providing a reference for battery thermal management.

2.3 Thermal Abuse Model

The thermal abuse model was used to study battery thermal safety, coupling internal exothermic reactions to simulate the occurrence and development of thermal runaway under thermal abuse. A review of abuse testing and simulation literature was conducted, and multiple exothermic reactions were selected to establish thermal models under abuse conditions such as hot box, short circuit, overcharge, and needle penetration. The role of fluorinated binders in thermal runaway was analyzed, and their influence was found to be relatively small.

Upgrading the one-dimensional thermal abuse model to a three-dimensional model, considering the shape, size, and material temperature distribution of battery components, and simulating oven experiments revealed that smaller batteries dissipate heat faster and are less prone to thermal runaway. A numerical simulation model of the nail penetration experiment, through electrochemical control equations and thermal abuse equations, accurately predicts temperature changes and the onset of thermal runaway during the nail penetration process, consistent with experimental results, thus solving the problem of time-consuming and expensive nail penetration experiments.

3. Conclusion and Outlook

Lithium-ion batteries, due to their excellent performance, are widely used in consumer electronics, power, and energy storage, but thermal safety issues hinder their widespread adoption. The core reason for thermal runaway is the inability to dissipate abnormal heat in a timely manner, leading to heat accumulation and a sudden temperature rise. Both experimental methods and model simulation methods are key tools for studying thermal problems, each with its advantages and disadvantages: experimental methods can accurately obtain heaat generation data under real-world conditions, but the process is complex, time-consuming, and expensive; model simulation methods are simple and have a short cycle, but they have certain errors and may deviate from reality.

Future research should organically combine these two approaches: using simulation results to guide experimental design, shortening experimental cycles and reducing budgets; and using experimental data to verify and revise simulation models, improving simulation accuracy. Through this synergy, we can delve deeper into the thermal characteristics of lithium-ion batteries, optimize thermal management solutions, and promote the safe, efficient, and large-scale application of lithium-ion batteries.