In recent years, the demand for high-energy-density and safe energy storage systems has surged, driving extensive research into advanced li ion battery technologies. Among these, hybrid solid-liquid electrolyte li ion batteries have emerged as a promising candidate, combining the benefits of both solid and liquid components to potentially overcome the limitations of traditional liquid electrolyte systems, such as safety hazards and interfacial issues. This work investigates the thermal runaway behavior of such hybrid systems, focusing on a configuration with LATP solid electrolyte paired with a commercial ester-based liquid electrolyte, using high-capacity NCM811 cathode and C@SiO anode materials. We employ adiabatic accelerating rate calorimetry and gas chromatography to simulate thermal runaway and analyze gas evolution across full state-of-charge (SOC) and various states-of-health (SOH) during the battery lifecycle. Additionally, ultrasonic testing is utilized to monitor changes in electrolyte content. Our findings reveal that thermal safety degrades with increasing SOC, while gas production rises with a higher proportion of combustible gases. Interestingly, as capacity fades to 70% retention, thermal stability improves, which we attribute to the conversion of liquid electrolyte into solid electrolyte interphases. This study provides foundational insights into the thermal runaway mechanisms of hybrid solid-liquid li ion batteries, supporting future design and safety enhancements.
The development of li ion batteries has revolutionized portable electronics and electric vehicles, yet safety concerns remain a critical barrier. Thermal runaway—a self-sustaining exothermic reaction leading to catastrophic failure—poses significant risks in conventional liquid electrolyte li ion batteries due to electrolyte flammability and electrode instability. Hybrid solid-liquid systems aim to mitigate these issues by incorporating solid electrolytes like LATP, which enhance mechanical stability and reduce leakage, while retaining some liquid electrolyte for improved ion transport. However, the thermal behavior of such hybrids under different operating conditions is not well-understood. Here, we systematically explore the thermal runaway characteristics as a function of SOC and SOH, using a comprehensive experimental approach. The goal is to correlate thermal parameters with electrochemical changes, offering guidance for safer li ion battery designs.
Our experimental methodology begins with the fabrication of hybrid solid-liquid li ion battery samples. We use an automated production line to assemble soft-pouch cells with dimensions of 7.5 cm × 6.4 cm × 0.5 cm, featuring NCM811 cathode, C@SiO anode, LATP-coated PP separator, and a matched commercial electrolyte. The cells are cycled between 2.8 V and 4.2 V, with a nominal capacity of 3.6 Ah at 0.2 C rate. Consistency is ensured through capacity screening and electrochemical impedance spectroscopy (EIS), where cells with similar charge-discharge profiles and impedance spectra are selected. SOC adjustment involves discharging cells to specific capacities (0%, 25%, 50%, 75%, and 100% SOC) based on a reference capacity, while SOH samples are obtained by cycling at 1 C rate until capacity retention reaches 90%, 80%, and 70% of the initial value. Thermal runaway tests are conducted in a modified adiabatic accelerating rate calorimeter, heating at 1.0 °C/min to trigger runaway, while monitoring temperature and pressure. Key parameters like self-heating onset temperature ($T_0$, where $dT/dt > 1.2$ °C/min), thermal runaway onset temperature ($T_c$, where $dT/dt > 2.0$ °C/min), maximum temperature ($T_{\text{max}}$), and maximum heating rate ($(dT/dt)_{\text{max}}$) are derived. Gas products are collected and analyzed via gas chromatography to quantify components such as CO$_2$, CO, H$_2$, CH$_4$, C$_2$H$_4$, C$_2$H$_6$, and C$_2$H$_2$. Ultrasonic scanning is performed in transmission mode to visualize electrolyte distribution changes.
The thermal runaway dynamics of li ion batteries can be modeled using kinetic equations. For instance, the self-heating rate often follows an Arrhenius relationship: $$ \frac{dT}{dt} = A \exp\left(-\frac{E_a}{RT}\right) $$ where $A$ is the pre-exponential factor, $E_a$ is the activation energy, $R$ is the gas constant, and $T$ is the temperature. In hybrid systems, the presence of both solid and liquid phases modifies this behavior, as reactions at interfaces contribute to heat generation. The overall heat release $Q$ during thermal runaway can be approximated by integrating the heat flow: $$ Q = \int_{T_0}^{T_{\text{max}}} C_p \, dT + \sum \Delta H_{\text{rxn}} $$ where $C_p$ is the heat capacity and $\Delta H_{\text{rxn}}$ represents enthalpy changes from decomposition reactions. For our li ion battery samples, we observe two exothermic peaks, corresponding to cathode decomposition and electrolyte breakdown, which align with reported mechanisms.
Results for fresh li ion battery samples at different SOC levels are summarized in Table 1. As SOC increases from 0% to 100%, $T_0$ decreases from 176.2 °C to 160.5 °C, and $T_c$ drops from 232.1 °C to 165.0 °C, indicating reduced thermal stability. $T_{\text{max}}$ rises dramatically from 276.1 °C to 720.7 °C, while $(dT/dt)_{\text{max}}$ escalates from 156 °C/min to 4722 °C/min. This trend underscores the heightened reactivity at high SOC, likely due to greater lithiation in the anode and delithiation in the cathode, which lowers the energy barrier for exothermic reactions. The pressure during thermal runaway also surges with SOC, from 5.17 bar to 17.38 bar, correlating with increased gas production. Gas volume grows from 67.54 mmol at 0% SOC to 452.39 mmol at 100% SOC, with a shift toward combustible species like CO and H$_2$. These findings emphasize the critical role of SOC management in enhancing the safety of li ion batteries.
| SOC (%) | $T_0$ (°C) | $T_c$ (°C) | $T_{\text{max}}$ (°C) | $(dT/dt)_{\text{max}}$ (°C/min) | Gas Volume (mmol) | Pressure Rise (bar) |
|---|---|---|---|---|---|---|
| 0 | 176.2 | 232.1 | 276.1 | 156 | 67.54 | 3.32 |
| 25 | 172.4 | 183.5 | 498.4 | 1716 | 189.22 | 8.45 |
| 50 | 170.7 | 177.1 | 552.1 | 2136 | 254.67 | 10.91 |
| 75 | 166.2 | 167.8 | 624.4 | 3240 | 338.75 | 13.24 |
| 100 | 160.5 | 165.0 | 720.7 | 4722 | 452.39 | 15.81 |
The gas composition analysis further elucidates the decomposition pathways. At 0% SOC, CO$_2$ dominates (85.29% of total gas), primarily from SEI layer decomposition and carbonate reactions. As SOC rises, the proportion of CO$_2$ decreases to 45.22% at 100% SOC, while combustible gases like CO and H$_2$ increase to 23.39% and 14.98%, respectively. This shift suggests enhanced reactions between lithium species and electrolytes at higher lithiation levels. The production of hydrocarbons (CH$_4$, C$_2$H$_4$, etc.) also grows, indicating complex decomposition of organic components. These trends highlight the interplay between electrochemical state and thermal runaway chemistry in li ion batteries, which must be considered for safety protocols.
Turning to cycle life effects, we examine li ion battery samples at different SOH levels (100%, 90%, 80%, and 70% capacity retention). Table 2 presents the thermal parameters. Interestingly, $T_0$ shows a slight dip at 90% SOH (169.4 °C) before recovering, while $T_c$ remains relatively stable around 223-232 °C. $T_{\text{max}}$ increases with cycling, from 276.1 °C at 100% SOH to 366.3 °C at 70% SOH, but $(dT/dt)_{\text{max}}$ peaks at 80% SOH (161.1 °C/min) and then declines to 147.3 °C/min at 70% SOH. This reduction in heating rate at lower SOH suggests improved thermal stability, possibly due to electrolyte consumption and solid electrolyte formation. Gas volume rises gradually from 67.54 mmol to 79.52 mmol as capacity fades, with CO$_2$ content increasing steadily, implying thicker SEI layers or more decomposition products.
| SOH (%) | $T_0$ (°C) | $T_c$ (°C) | $T_{\text{max}}$ (°C) | $(dT/dt)_{\text{max}}$ (°C/min) | Gas Volume (mmol) | Weight Loss (%) |
|---|---|---|---|---|---|---|
| 100 | 176.2 | 232.1 | 276.1 | 156.0 | 67.54 | 13.58 |
| 90 | 169.4 | 223.9 | 309.5 | 150.9 | 67.14 | 14.22 |
| 80 | 177.2 | 224.2 | 329.0 | 161.1 | 74.70 | 15.87 |
| 70 | 177.4 | 222.7 | 366.3 | 147.3 | 79.52 | 16.54 |
Ultrasonic imaging corroborates these observations. Fresh li ion battery samples show uniform electrolyte distribution (yellow hues in scans), but as cycling progresses, signal attenuation increases (green to blue shifts), indicating reduced liquid electrolyte content. At 70% SOH, high-voltage scans reveal sparse electrolyte pockets, predominantly at edges. This visual evidence supports the hypothesis that liquid electrolyte gradually transforms into solid-like phases through side reactions, altering the thermal response. The kinetic implications can be expressed via a modified degradation model: $$ \frac{dC}{dt} = -k C^n $$ where $C$ is electrolyte concentration, $k$ is a rate constant, and $n$ is the reaction order. For our li ion batteries, $n$ may approach 1, suggesting first-order consumption during cycling.
To quantify gas composition changes, we analyze the molar fractions of key species. Table 3 details the gas components at different SOHs (0% SOC). CO$_2$ remains the major product, increasing from 57.40 mmol (85.29%) at 100% SOH to 65.88 mmol (82.84%) at 70% SOH. Combustible gases like H$_2$ and CO exhibit non-monotonic trends: H$_2$ peaks at 6.75 mmol (9.04%) at 80% SOH, while CO is lowest at 4.17 mmol (5.58%) at 80% SOH. This variability points to competing reactions—such as electrolyte reduction and binder decomposition—that evolve with cycle life. The overall gas yield $G$ can be related to capacity loss $\Delta Q$ through an empirical equation: $$ G = \alpha \Delta Q + \beta $$ where $\alpha$ and $\beta$ are constants derived from our data, reflecting the gas generation per unit capacity fade in li ion batteries.
| Gas Species | 100% SOH (mmol, %) | 90% SOH (mmol, %) | 80% SOH (mmol, %) | 70% SOH (mmol, %) |
|---|---|---|---|---|
| CO$_2$ | 57.40, 85.29% | 58.12, 86.58% | 62.34, 83.47% | 65.88, 82.84% |
| CO | 3.89, 5.78% | 3.45, 5.14% | 4.17, 5.58% | 5.12, 6.44% |
| H$_2$ | 3.45, 5.13% | 3.22, 4.80% | 6.75, 9.04% | 5.89, 7.41% |
| CH$_4$ | 0.89, 1.32% | 0.95, 1.42% | 1.12, 1.50% | 1.34, 1.68% |
| C$_2$H$_4$ | 0.67, 1.00% | 0.71, 1.06% | 0.89, 1.19% | 0.92, 1.16% |
| C$_2$H$_6$ | 0.45, 0.67% | 0.48, 0.72% | 0.56, 0.75% | 0.61, 0.77% |
| C$_2$H$_2$ | 0.22, 0.33% | 0.25, 0.37% | 0.31, 0.42% | 0.35, 0.44% |
Discussion of these results centers on the mechanisms driving thermal runaway in hybrid solid-liquid li ion batteries. At high SOC, the increased lithiation degree in the anode enhances reactivity with the electrolyte, leading to vigorous exothermic reactions. The decomposition of NCM811 cathode releases oxygen, which further oxidizes electrolyte components, producing CO$_2$ and heat. This can be modeled using a coupled reaction set: $$ \text{Li}_x\text{C} + \text{Electrolyte} \rightarrow \text{Li}_2\text{CO}_3 + \text{Gases} + \text{Heat} $$ $$ \text{LiNi}_{0.8}\text{Co}_{0.1}\text{Mn}_{0.1}\text{O}_2 \rightarrow \text{Delithiated Oxide} + \text{O}_2 + \text{Heat} $$ The oxygen then participates in combustion-like processes, amplifying thermal runaway. As for cycle life effects, the gradual consumption of liquid electrolyte reduces the fuel available for exothermic reactions, thereby moderating the heating rate at low SOH. This is consistent with ultrasonic data showing electrolyte depletion. Moreover, the formation of solid electrolyte interphases (SEI) and possible conversion of liquid into solid phases may create a more thermally stable matrix, delaying runaway onset. However, the rising $T_{\text{max}}$ suggests that once triggered, the reactions become more intense due to accumulated degradation products.
The implications for li ion battery safety are profound. Our study indicates that managing SOC and monitoring SOH can help mitigate thermal risks. For instance, operating li ion batteries at lower SOC levels may enhance safety, albeit at the cost of energy density. Similarly, early detection of electrolyte loss via ultrasonic or impedance techniques could serve as a warning for impending instability. From a design perspective, optimizing the solid-to-liquid ratio in hybrid electrolytes might balance performance and safety. We propose a safety index $S$ based on our parameters: $$ S = \frac{T_c \cdot (dT/dt)_{\text{max}}^{-1}}{G} $$ where higher $S$ denotes better safety. For our li ion batteries, $S$ decreases with SOC but shows a rebound at 70% SOH, aligning with observed improvements.
In conclusion, we have systematically investigated the thermal runaway behavior of hybrid solid-liquid li ion batteries across SOC and SOH ranges. Fresh li ion batteries exhibit reduced thermal stability at higher SOC, with increased gas production and combustible gas fractions. Cycling leads to capacity fade but can improve thermal stability at 70% retention, likely due to electrolyte solidification. These insights advance the understanding of failure mechanisms in advanced li ion batteries, paving the way for safer energy storage solutions. Future work should focus on real-time monitoring and material innovations to further suppress thermal runaway in li ion batteries.
To summarize key equations and relationships, the thermal kinetics of li ion batteries can be described by: $$ \frac{dT}{dt} = f(\text{SOC}, \text{SOH}) \cdot A \exp\left(-\frac{E_a}{RT}\right) $$ where $f(\text{SOC}, \text{SOH})$ is a function accounting for state-dependent reactivity. The gas evolution model: $$ G = \sum_i k_i [\text{Reactant}_i] $$ with $k_i$ as rate constants for various decomposition pathways. These models, combined with our empirical data, provide a framework for predicting and enhancing the safety of li ion batteries in diverse applications.