In this thesis, the thermal safety and thermal management technology of retired ternary lithium-ion vehicle traction batteries were investigated from both cell level and module level. First, the residual usability and inconsistency of retired Li(Ni₀.₅Co₀.₂Mn₀.₃)O₂/C (NCM523), Li(Ni₀.₆Co₀.₂Mn₀.₂)O₂/C (NCM622), and Li(Ni₀.₈Co₀.₁Mn₀.₁)O₂/C (NCM811) 18650 cells were systematically inspected. Then, the electro-thermal behavior under adiabatic conditions and thermal runaway (TR) characteristics triggered by overheating abuse were compared in terms of temperature rise, voltage response and microstructure evolution. In parallel, a high-insulation and high-thermal-conductivity composite phase change material (CPCM) composed of paraffin, expanded graphite, silicon nitride and epoxy resin was developed, and the optimal formulation was selected through comprehensive characterization. The CPCM was eventually integrated into a retired NCM811 module for passive thermal management. The results show that the NCM811 cells exhibit the highest remaining capacity but the most severe heat generation and the lowest thermal runaway onset temperature, which is consistent with the highest irreversible damage to the anode and cathode active particles. The obtained Si₃N₄-based CPCM with 8 wt% Si₃N₄ presents a volume resistivity of 8.93×10¹¹ Ω·cm, a thermal conductivity of 2.39 W·m⁻¹·K⁻¹ and a leakage rate below 0.2% after high-temperature cycled exudation tests. When applied to the retired NCM811 module, the CPCM-5 cooling system maintains the maximum temperature below 48.2 °C at a 2.0C discharge rate, and the maximum temperature difference is lower than 1.0 °C. Therefore, the designed high-insulation high-thermal-conductivity CPCM is promising for the thermal management of high-energy-density vehicle traction batteries.
1. Introduction
In response to the strategic demand of green development, electric vehicles (EVs) have experienced explosive growth. Simultaneously, the number of vehicle traction batteries retired from EVs is expanding sharply. However, these retired batteries still possess considerable utilization value, especially retired ternary lithium-ion batteries with high energy density. Because of their internal material structure, however, the risk of thermal runaway is significantly higher than that of low-nickel or lithium iron phosphate batteries. Thus, investigating the thermal safety characteristics and developing effective thermal management systems for retired ternary vehicle traction batteries is of great importance.
Lithium-ion batteries (LIBs) are widely used in vehicle traction applications due to their high specific energy, long cycle life, low self-discharge rate and no memory effect. The energy density of LIBs is continuously improved by increasing the nickel content in the layered oxide cathode, i.e., from NCM111 to NCM523, NCM622, NCM811 and even NCA. Nevertheless, the increase of nickel content is accompanied by a decline in thermal stability. Under thermal abuse conditions, the internal chemical reactions can cascade into thermal runaway (TR), which is regarded as one of the most severe safety hazards for vehicle traction batteries. Many fire accidents reported for EVs have been attributed to TR triggered by internal short circuits, overcharging, mechanical abuse or overheating.
The heat generation of a working vehicle traction battery arises mainly from the irreversible Joule heat, polarization heat, and reversible entropic heat. When the heat removal capability is insufficient, the cell temperature rises. Once the temperature exceeds a critical value, a series of exothermic side reactions occur, including decomposition of the solid electrolyte interface (SEI), reaction between the anode and electrolyte, decomposition of the electrolyte, decomposition of the cathode active material, and oxidation of the combustible gases. These reactions release a large amount of heat and accelerate the temperature rise, eventually causing TR. It is therefore necessary to systematically study the TR behavior of different nickel-content cathode materials and to build an efficient module-level heat dissipation strategy.
Among the thermal management techniques for vehicle traction batteries, phase change material (PCM) cooling has received increasing attention because of its passive operation, high heat-absorption density, uniform temperature distribution and low parasitic energy consumption. However, the intrinsic low thermal conductivity and potential leakage of pure paraffin restrict its practical application. In this work, a thermally conductive and electrically insulating skeleton composed of silicon nitride (Si₃N₄) and expanded graphite (EG) is introduced into paraffin/epoxy, aiming at solving the trade-off between high thermal conductivity and high electrical insulation. The obtained composite phase change material is subsequently validated in a retired NCM811 cylindrical-cell module.
The main contributions of this work are summarized as follows:
- We conducted a comparative study of the reusability and inconsistency of retired NCM523, NCM622 and NCM811 vehicle traction batteries through systematic screening.
- We revealed the thermal-runaway characteristics of these batteries from both macroscopic temperature profiles and microscopic electrode material damage.
- We designed a Si₃N₄/EG hybrid thermally conductive skeleton and obtained a novel CPCM with high thermal conductivity, high volume resistivity, good mechanical strength and low leakage.
- We applied the optimized CPCM to a retired NCM811 battery module and demonstrated its temperature-control and temperature-uniformity capability under continuous charge–discharge cycling.
2. Experimental
2.1 Battery samples and screening
The experimental objects were three types of commercial 18650 ternary lithium-ion vehicle traction batteries with different nickel content. The nominal specifications are given in Table 1.
| Parameter | NCM523 | NCM622 | NCM811 |
|---|---|---|---|
| Cathode material | LiNi₀.₅Co₀.₂Mn₀.₃O₂ | LiNi₀.₆Co₀.₂Mn₀.₂O₂ | LiNi₀.₈Co₀.₁Mn₀.₁O₂ |
| Nominal voltage (V) | 3.6 | 3.6 | 3.6 |
| Nominal capacity (Ah) | 2.60 | 2.75 | 3.05 |
| Charge cut-off voltage (V) | 4.2 | 4.2 | 4.2 |
| Discharge cut-off voltage (V) | 2.75 | 2.5 | 2.5 |
| Mass (g) | 46.5 | 49 | 47 |
| Mass energy density (Wh/kg) | 180 | 206 | 234 |
A total of 150 retired cells, 50 of each type, were collected from a retired EV battery pack after capacity fade to approximately 80% of the rated value. The screening procedure included visual inspection, open-circuit voltage (OCV) measurement, internal alternating-current resistance (ACIR) measurement and capacity test. The internal resistance was measured using a multi-channel battery tester at 1 kHz under the fully-charged state. The capacity was measured using a program of three consecutive charge/discharge cycles, and the final discharge capacity of the third cycle was adopted.
2.2 Charge–discharge and temperature measurement
Charge–discharge tests were performed on a Neware battery test station (CT-6008n-60 V/100 A) inside a temperature-controlled chamber. A T-type thermocouple (accuracy ±0.1 °C) was attached to the middle of the cell surface, and the temperature was recorded by an Agilent 34970A data acquisition instrument at intervals of 1 s. The cell was fully charged by a constant-current constant-voltage (CC-CV) protocol, i.e., 1.0C constant current to 4.2 V followed by a constant-voltage step until the current decreased to 31 mA. After resting for 30 min, the cell was discharged at different current rates (0.5C, 1.0C and 2.0C) in an adiabatic environment.
2.3 Adiabatic calorimetry and thermal runaway test
The adiabatic tests were conducted in an EV+ Accelerating Rate Calorimeter (EV-ARC). The heat-wait-seek (H-W-S) mode was used to detect the onset temperature of self-heating. The criterion of self-heating was set as 0.03 °C·min⁻¹. The ARC heated the cell in steps of 2 °C, and once self-heating was detected, the heaters were controlled to provide an adiabatic environment so that all heat generated by the cell remained inside the cell.
For the thermal-runaway-triggering experiments, fully charged cells with 100% SOC were heated by an external heating rod located at the same geometry as a cylindrical cell to simulate a thermal abuse condition in a module. The cell voltage and surface temperature were synchronously recorded until thermal runaway was complete.
2.4 Materials characterization
After the TR tests, the battery cans were opened inside an argon-filled glove box. The anode and cathode sheets were separated and gently rinsed with dimethyl carbonate to remove the residual electrolyte. Trace amounts of active powders were collected and dried. The surface morphologies of the active particles before and after TR were observed using a scanning electron microscope (SEM, Hitachi S-3400N-II).
2.5 Preparation of Si₃N₄-based composite phase change material
Paraffin wax (PA) with a melting point of 48.8 °C and latent heat of 225.8 J·g⁻¹ was used as the phase-change component. Expanded graphite (EG) was fabricated by expanding expandable graphite at 700 °C for 1 min. Epoxy resin (ER) composed of component A and curing agent B (mass ratio 1:1) was adopted as the anti-leakage matrix. Si₃N₄ particles with an average particle size of 2 μm served as the high-thermal-conductivity insulating filler.
A melt-mixing and gelation method was used to prepare the CPCM. The experimental procedure is shown in Figure… (not repeated here). First, paraffin was melted in a water bath at 70 °C. Then EG was added and stirred for 60 min. Subsequently Si₃N₄ powders were dispersed into the mixture and stirring continued for 15 min. Finally, the epoxy resin and curing agent were added, and the mixture was poured into acrylic moulds and cured at 90 °C for 10 h. Six samples with different Si₃N₄ contents are listed in Table 2.
| Sample | PA (mass fraction) | EG (mass fraction) | ER (mass fraction) | Si₃N₄ (mass fraction) |
|---|---|---|---|---|
| CPCM-1 | 83.5% | 1.5% | 15% | 0 |
| CPCM-2 | 81.5% | 1.5% | 15% | 2% |
| CPCM-3 | 79.5% | 1.5% | 15% | 4% |
| CPCM-4 | 77.5% | 1.5% | 15% | 6% |
| CPCM-5 | 75.5% | 1.5% | 15% | 8% |
| CPCM-6 | 73.5% | 1.5% | 15% | 10% |
2.6 Characterization of CPCM
Thermogravimetric analysis (TGA) was performed under N₂ atmosphere from 30 to 800 °C at a heating rate of 10 °C·min⁻¹. Differential scanning calorimetry (DSC) was used to measure the phase-change temperature and latent heat from 30 to 80 °C. X-ray diffraction (XRD) was conducted in the 2θ range from 10° to 70°. Thermal conductivity was measured by a Hot Disk thermal constant analyzer at 50 °C. The volume resistivity was measured with an ultra-high-resistance meter on disc samples. For anti-leakage performance, cylindrical samples (Φ20 × 10 mm) were placed on a filter paper in an oven at 70 °C for 12 h and weighed every 1.5 h. The leakage rate is calculated by:
$$ K_{\mathrm{leakage}} = \frac{M_0 – M_1}{M_0} \times 100\% $$
where \(M_0\) is the initial sample mass and \(M_1\) is the mass after each weighing.
2.7 Module-level thermal management measurement
Retired NCM811 cells were selected to build 6P1S modules (six cells in parallel, one row). The module capacity was approximately 18.3 Ah. Three module configurations were prepared: (i) no heat dissipation (natural air convection), (ii) module filled with CPCM-1, and (iii) module filled with CPCM-5. The battery cells were placed in an acrylic case and molten CPCM was poured around the cells. After solidification, the module was exposed to air. Four T-type thermocouples were attached to the upper, middle and lower positions of the cell surface. The cyclic test was conducted at 25 °C in a thermostatic chamber. The cells were charged at 1.0C to 4.2 V followed by constant-voltage charging until 0.03 A, rested for 60 min, and then discharged at 2.0C. Ten repeated cycles were carried out to verify the long-term stability.
3. Results and Discussion
3.1 Screening results of retired vehicle traction batteries
Through the initial visual inspection, 10 cells failed due to shell cracking, rusting or electrolyte leakage. A further 7 cells failed the voltage test because the open-circuit voltage was lower than 2 V or zero. In addition, 3 cells were excluded due to abnormal internal resistance. Thus, 130 of the 150 retired ternary vehicle traction batteries were considered qualified, accounting for 86.7%. The failure distributions are given in Table 3.
| Failure reason | NCM523 | NCM622 | NCM811 | Total | Percentage |
|---|---|---|---|---|---|
| Appearance defect | 2 | 3 | 5 | 10 | 6.7% |
| Abnormal voltage | 2 | 2 | 3 | 7 | 4.7% |
| High internal resistance or voltage anomaly | 1 | 1 | 1 | 3 | 2.0% |
| Qualified | 45 | 44 | 41 | 130 | 86.7% |
After screening, the capacities of the qualified NCM523, NCM622 and NCM811 cells were measured, as shown in Figure. The average values were 2414.89 mAh, 2711.16 mAh and 2790.15 mAh, respectively. The NCM811 cells had an average capacity 15.5% higher than that of NCM523 and 2.9% higher than that of NCM622. The capacity standard deviations were relatively small, indicating good consistency after screening. The internal resistance values showed a similar trend, with the average resistance of the qualified cells being 22.62 mΩ, 22.45 mΩ and 22.33 mΩ for NCM523, NCM622 and NCM811, respectively.
The retention and recovery rates were measured after storage at room temperature for 30 days. The capacity retention rate \(\eta_{\mathrm{retention}}\) is defined as:
$$ \eta_{\mathrm{retention}} = \frac{C_3}{C_2} \times 100\% $$
and the capacity recovery rate \(\eta_{\mathrm{recovery}}\) is defined as:
$$ \eta_{\mathrm{recovery}} = \frac{C_2 – C_0}{C_1 – C_0} \times 100\% $$
where \(C_0\) is the fully discharged energy, \(C_1\) is the charged energy after storage, \(C_2\) is the discharged energy after completing a full charge/discharge cycle, and \(C_3\) is the remaining energy after 30 days of storage. The statistics show that 90.8% of the retired cells retained more than 95% of their stored capacity, and 79.0% of the cells had a recovery rate above 90%. This verifies the high reusability of retired ternary vehicle traction batteries for second-life applications.
3.2 Temperature rise under adiabatic discharge
Figures present the surface-temperature evolution of the three nickel-content cells during constant-current discharges in an adiabatic chamber at 25 °C. At a low discharge rate of 0.5C, the maximum temperatures of NCM523, NCM622 and NCM811 were 37.0 °C, 39.0 °C and 34.9 °C, respectively, with no evident difference. At 1.0C, the maximum temperatures were 42.9 °C, 47.3 °C and 45.4 °C. However, at the high discharge rate of 2.0C, the NCM811 cell reached 70.9 °C, while NCM523 and NCM622 only reached 56.7 °C and 68.8 °C, respectively. The average temperature rise rates at 2.0C are compared in Table 4.
| Cell type | Discharge time (min) | Maximum temperature (°C) | Temperature rise rate (°C/min) |
|---|---|---|---|
| NCM523 | 30.9 | 56.7 | 1.026 |
| NCM622 | 30.5 | 68.8 | 1.436 |
| NCM811 | 29.5 | 70.9 | 1.556 |
The similar phenomenon was observed at an ambient temperature of 55 °C. The NCM811 cell exhibited a maximum temperature of 96.6 °C and a temperature-rise rate of 1.4597 °C·min⁻¹ at 2.0C discharge, which was 1.92 times and 1.20 times that of NCM523 and NCM622, respectively. The results imply that the heat-generation rate of high-nickel vehicle traction batteries is significantly larger than that of low-nickel batteries under high-rate discharge, confirming the more severe thermal risk of NCM811.
3.3 Thermal runaway behavior
Thermal runaway experiments were carried out under adiabatic external heating conditions. Figure … illustrates the voltage and temperature profiles for NCM523, NCM622 and NCM811. Three characteristic temperatures are defined: \(T_1\) is the self-heating onset temperature when the temperature rise rate exceeds 0.03 °C·min⁻¹; \(T_2\) is the TR trigger temperature when the temperature rise rate exceeds 1 °C·s⁻¹; \(T_3\) is the maximum temperature reached during TR. Table 5 lists the measured characteristic temperatures.
| Cell type | T₁ (°C) | T₂ (°C) | T₃ (°C) | Maximum temperature rise rate (°C/min) |
|---|---|---|---|---|
| NCM523 | 81.3 | 184.3 | 574.94 | 4264.71 |
| NCM622 | 80.8 | 171.7 | 670.63 | 9069.1 |
| NCM811 | 78.4 | 139.9 | 676.44 | 9420.33 |
It can be seen that the onset self-heating temperature \(T_1\) declines slightly with increasing nickel content, whereas the TR-trigger temperature \(T_2\) of NCM811 is 44.4 °C lower than that of NCM523. Moreover, NCM811 simultaneously possesses the highest peak temperature \(T_3\) and the fastest temperature-rise rate, indicating a more catastrophic consequence once thermal runaway is triggered. The reduced thermal stability is attributed to the increased content of Ni⁴⁺ in the delithiated state, which readily decomposes into NiO and releases oxygen. The released oxygen then oxidizes the organic solvents and accelerates the exothermic chain reactions.
3.4 Post-thermal-runaway morphology and microstructure
After TR, the appearance of the cells was severely damaged, and the degree of damage increased with nickel content. For NCM811, the top cap was deformed and the inner active materials leaked seriously. The anode current collector became brittle and fragmented; some areas completely lost the active coating. SEM images of the anode active particles before and after TR are compared in Table 6. It is clearly shown that TR causes the anode particles to lose their smooth morphology and produce fissures and agglomerated debris. Among the three types, NCM811 shows the most severe particle destruction, while NCM523 remains relatively complete. This microstructural deterioration leads to a decrease in the active surface area, an increase in the internal resistance, and an irreversible capacity loss, which is directly responsible for the degraded electrochemical performance of the battery after experiencing a thermal event.
| Condition | NCM523 | NCM622 | NCM811 |
|---|---|---|---|
| Before TR | Uniform, smooth particles | Uniform, smooth particles | Uniform, smooth particles |
| After TR | Slight cracking, minor damage | Severe cracking and detachment | Complete fragmentation and severe agglomeration |
These observations strongly support the conclusion that the thermal-electrochemical deterioration of high-nickel vehicle traction batteries originates from the structural collapse of active materials under thermal abuse. Thus, an efficient battery thermal management system is indispensable to keep the battery working below the self-heating onset temperature, especially for high-nickel cells.
4. Performance of Si₃N₄-based Composite Phase Change Materials
4.1 Thermal stability and latent-heat characteristics
The TGA results of the CPCMs show that pure paraffin begins to decompose near 200 °C, while the introduction of Si₃N₄ does not significantly shift the decomposition process but reduces the total mass loss at elevated temperatures because of the inorganic residue. At 800 °C, the residual mass corresponds approximately to the content of Si₃N₄ and inorganic additives, confirming the effective incorporation of Si₃N₄ into the composite. All the CPCMs have two weight-loss regions: the first between 30 and 200 °C is mainly caused by evaporation and decomposition of some low-molecular-weight components, and the second between 300 and 400 °C corresponds to the complete decomposition of paraffin and the epoxy network.
DSC curves illustrate the melting behavior of the prepared CPCMs. The phase-change temperatures remain between 47 and 48 °C, which is close to the phase-transition point of pure paraffin but is more suitable for the operating temperature window of vehicle traction batteries. The latent heat values decrease almost linearly with the reduction of the paraffin mass fraction. For CPCM-5 with 8 wt% Si₃N₄, the measured latent heat is 147.14 J·g⁻¹. Since the paraffin content in CPCM-5 is 75.5%, the theoretical latent heat would be \(0.755 \times 225.8 = 170.5\) J·g⁻¹. The slightly lower experimental value is attributed to the physical adsorption of the porous EG and the partial loss of paraffin during the moulding process. However, a latent heat of ~147 J·g⁻¹ is sufficient for passive thermal buffering.
4.2 Thermal conductivity and electrical insulation
Table 7 summarizes the thermal conductivity and volume resistivity of the CPCMs with different Si₃N₄ contents.
| Sample | Si₃N₄ content (wt%) | Thermal conductivity (W·m⁻¹·K⁻¹) | Volume resistivity (×10¹¹ Ω·cm) |
|---|---|---|---|
| CPCM-1 | 0 | 1.11 | 3.37 |
| CPCM-2 | 2 | 1.30 | 4.95 |
| CPCM-3 | 4 | 1.60 | 5.95 |
| CPCM-4 | 6 | 2.05 | 6.95 |
| CPCM-5 | 8 | 2.39 | 8.93 |
| CPCM-6 | 10 | 2.53 | 6.44 |
From the data we can see that the thermal conductivity increases steadily with the Si₃N₄ content because more Si₃N₄ particles form a more continuous three-dimensional heat-conduction network together with EG. At the same time, the volume resistivity first increases and then decreases. When the Si₃N₄ content reaches 8 wt%, the volume resistivity peaks at 8.93×10¹¹ Ω·cm, which is far above the conventional insulation criterion of 1×10⁹ Ω·cm. The resistivity decline at 10 wt% can be explained by two factors: first, excessive Si₃N₄ particles tend to aggregate, producing interfacial defects that facilitate charge hopping; second, the interface polarization between Si₃N₄ and the polymer matrix may become stronger at high filler loadings.
4.3 Mechanical properties and anti-leakage performance
The mechanical strength of the CPCMs was evaluated by tensile and three-point bending tests. As shown in Table 8, both tensile strength and flexural strength initially increase with Si₃N₄ content, reaching their maximum values at 8 wt%. Compared with CPCM-1, the tensile strength and flexural strength of CPCM-5 increase by 79.9% and 43.6%, respectively. The improvement is ascribed to the stiff Si₃N₄ particles filling the micro-voids and the strong interfacial adhesion between Si₃N₄ and the polymer matrix. A further increase to 10 wt% leads to a slight decrease in both strengths, likely due to the agglomeration of Si₃N₄ and the consequent stress-concentration points.
| Sample | Tensile strength (MPa) | Flexural strength (MPa) |
|---|---|---|
| CPCM-1 | 0.55 | 1.21 |
| CPCM-2 | 0.65 | 1.38 |
| CPCM-3 | 0.78 | 1.55 |
| CPCM-4 | 0.88 | 1.67 |
| CPCM-5 | 0.99 | 1.74 |
| CPCM-6 | 0.92 | 1.63 |
The high-temperature exudation tests show that all prepared CPCMs have a very low leakage rate lower than 0.2% after being held at 70 °C for 12 h. With the increase of Si₃N₄ content, the leakage rate slightly drops from 0.18% for CPCM-2 to 0.13% for CPCM-6. This is because more Si₃N₄ particles create a more tortuous path for the molten paraffin and enhance the capillary force in the composite network. Therefore, the anti-leakage performance is suitable for long-term vehicle traction battery operation.
4.4 XRD and SEM analysis
XRD patterns of the CPCMs display peaks characteristic of paraffin at \(2\theta = 21.52^\circ\) and \(24.51^\circ\), a graphite characteristic peak at \(26.77^\circ\), and Si₃N₄ peaks at \(36.14^\circ\), \(40.7^\circ\), and \(47.76^\circ\). No new diffraction peaks are observed, indicating that all components are physically mixed without chemical reaction. The SEM images show a smooth two-dimensional layered structure for paraffin and a worm-like porous structure for EG. The Si₃N₄ particles with submicron size are uniformly embedded in the network at the appropriate concentration. At 8 wt% Si₃N₄, the composite surface becomes dense and almost free of macro-cracks. However, at 10 wt% Si₃N₄, agglomerated Si₃N₄ clusters are clearly observed, which explains the slight deterioration of mechanical and electrical properties.
5. Application of CPCM in Thermal Management of Retired Vehicle Traction Battery Modules
5.1 Single-cell and module-level temperature control
In this section, the optimized CPCM-5 and the reference CPCM-1 were integrated into a battery thermal management system (BTMS) for retired 18650 NCM811 vehicle traction batteries. A natural-air-convection module was used as a baseline. The modules were discharged at 0.5C, 1.0C and 2.0C in a 25 °C environment. The peak temperatures of the three configurations are listed in Table 9.
| Discharge rate | Air cooling max. temp. (°C) | CPCM-1 cooling max. temp. (°C) | CPCM-5 cooling max. temp. (°C) |
|---|---|---|---|
| 0.5C | 36.1 | 34.5 | 33.9 |
| 1.0C | 48.9 | 43.0 | 41.6 |
| 2.0C | 67.7 | 52.0 | 48.2 |
Compared with natural air convection, the CPCM-5 module reduces the peak temperature by 6.1%, 14.9% and 28.8% at discharge rates of 0.5C, 1.0C and 2.0C, respectively. The better performance at a high rate is attributed to the synergy of the latent-heat absorption of paraffin and the fast heat spreading through the Si₃N₄/EG thermally conductive network.
The maximum temperature differences inside the modules are shown in Table 10. At the 2.0C discharge rate, the maximum in-module temperature difference of the air-cooled module reaches 3.6 °C, while CPCM-1 keeps it at 2.4 °C and CPCM-5 further reduces it to 1.3 °C. The excellent temperature-uniformity of CPCM-5 is also reflected by the fact that the upper-, middle- and lower-surface temperature curves are almost overlapping during discharge. This mitigates the local over-heating generated by the current collectors and cell tabs, thereby enhancing the safety and lifespan of the retired vehicle traction battery module.
| Discharge rate | Air cooling ΔTmax (°C) | CPCM-1 ΔTmax (°C) | CPCM-5 ΔTmax (°C) |
|---|---|---|---|
| 0.5C | 0.31 | 0.23 | 0.15 |
| 1.0C | 1.24 | 0.56 | 0.37 |
| 2.0C | 3.6 | 2.4 | 1.3 |
5.2 Cyclic charge–discharge stability
To evaluate the long-term cooling performance under realistic operation, the modules were subjected to ten consecutive cycles of 1.0C charge and 2.0C discharge. The natural-air-cooled module experienced a continuous temperature rise cycle after cycle and the maximum temperature eventually reached 69.5 °C. The CPCM-1 module stabilized at about 56.7 °C, whereas the CPCM-5 module maintained its maximum temperature at approximately 48.8 °C, and the difference between consecutive cycles was less than 0.7 °C. This indicates that the excellent thermal conductivity and phase-change enthalpy of CPCM-5 help to avoid heat accumulation and retain the cell temperature inside the optimal operating window of vehicle traction batteries.
6. Conclusion
In this thesis, the thermal-electrochemical characteristics of retired ternary vehicle traction batteries with different nickel contents were systematically studied, and a novel Si₃N₄-based high-insulation high-thermal-conductivity composite phase change material was developed for module-level thermal management. The main conclusions are as follows:
- After a multi-level screening of 150 retired cells, 130 cells (86.7%) were qualified as reusable. The NCM811 cells had the highest average discharge capacity (2790.15 mAh), while their surface temperature during discharge was also the highest among the three types. The capacity retention and recovery measurements indicated that most of the retired cells are suitable for second-life vehicle traction battery applications.
- With increasing nickel content in the cathode, the heat generation of the cell under high-rate discharge increases. NCM811 shows the highest maximum temperature and the highest temperature rise rate under adiabatic conditions at 2.0C discharge. In the overheating-triggered TR test, NCM811 has the lowest TR onset trigger temperature (\(T_2 = 139.9\) °C) but the highest peak temperature (\(T_3 = 676.44\) °C) and the fastest temperature rise rate (9420.33 °C·min⁻¹). The post-TR microstructure analysis confirms that the active material damage in the anode and the detachment from the copper current collector are most severe for NCM811, which accounts for its irreversible capacity fade and increased thermal sensitivity.
- To address the low thermal conductivity of paraffin and the conflict between high thermal conductivity and electrical insulation in composite phase change materials, we fabricated a Si₃N₄/EG hybrid thermally conductive network. The optimal formulation contains 8 wt% Si₃N₄, which provides a thermal conductivity of 2.39 W·m⁻¹·K⁻¹, a volume resistivity of 8.93×10¹¹ Ω·cm, a latent heat of 147.14 J·g⁻¹, a leakage rate below 0.2%, and simultaneously the maximum tensile and flexural strengths among all samples.
- The module-level experiment shows that the CPCM-5-based thermal management system is effective in suppressing the temperature rise of retired NCM811 modules; at 2.0C discharge the peak temperature is reduced from 67.7 °C (natural air cooling) to 48.2 °C, and the maximum module temperature difference is reduced to 1.3 °C. Under continuous cycling, the temperature remains stable, confirming the practical feasibility of the synthesized CPCM for secondary-use high-nickel vehicle traction batteries.
Future research will focus on improving the flame retardancy of the CPCM without sacrificing the latent heat and mechanical properties, and on exploring more reliable battery thermal runaway propagation inhibition strategies for large-scale retired vehicle traction battery energy storage systems.