In the context of global energy structure transformation, energy storage technology has become a critical enabler for mitigating the intermittency and volatility of renewable energy sources. Large-capacity energy storage cells are increasingly favored in modern energy storage systems and electric vehicles due to their high energy density, which reduces the cost per kilowatt-hour and simplifies system integration. However, the thermal safety and long-term reliability of these large-format cells remain significant challenges. During charge and discharge cycles, energy conversion inevitably generates heat, and if not properly managed, this heat accumulation can lead to accelerated degradation, thermal runaway, and catastrophic failures. In this study, we systematically investigate the heat generation behavior and capacity fading mechanisms of a 280 Ah lithium iron phosphate (LiFePO4) prismatic energy storage cell under various operating conditions. Our experimental approach focuses on the effects of discharge rate, ambient temperature, and spatial temperature distribution on thermal performance, as well as the correlation between temperature and capacity loss over cycling. Through detailed thermal measurements and electrochemical analysis, we aim to provide a comprehensive understanding of the thermal behavior of large-capacity energy storage cells and to offer guidance for the design of battery thermal management systems (BTMS) and battery management systems (BMS).
1. Experimental Methods
We selected a commercial 280 Ah prismatic LiFePO4/graphite cell (GSP71173204F, Penghui Energy) as the test subject. The key parameters of the tested energy storage cell are summarized in Table 1. All experiments were conducted using a Neware CT-4002-5V200A battery tester for precise current/voltage control, a Ruikai R-TD-00RF temperature and humidity chamber to maintain stable ambient temperatures, a TCP-XL multichannel temperature recorder for surface temperature monitoring, and a DF133-HFM-8 heat flux meter for measuring thermal flux. Eight thermocouples (T1 to T8) were attached to the cell surface at specific locations: T1 and T2 near the negative terminal, T3 and T7 at the center of the large face, T4, T5, and T8 near the bottom, and T6 near the positive terminal. This arrangement allowed us to capture the spatial temperature gradient across the large-format energy storage cell.
| Parameter | Value |
|---|---|
| Rated capacity (Ah) | 280 |
| Mass (kg) | 5.34 |
| Nominal voltage (V) | 3.2 |
| Charge cutoff voltage (V) | 3.65 |
| Discharge cutoff voltage (V) | 2.0 |
| Dimensions (mm × mm × mm) | 173 × 71 × 204 |
Before each test series, we performed an initial capacity calibration at 25 °C using a constant-current constant-voltage (CC-CV) protocol: charge at 1/3C to 3.65 V, then constant voltage until current dropped below 10 A, rest 4 h, then discharge at 1/3C to 2.0 V. The average of three cycles was taken as the nominal capacity. The internal resistance of the energy storage cell was measured using the hybrid pulse power characterization (HPPC) method at various states of charge (SOC). A 10 s discharge pulse followed by a 30 s rest and a 10 s charge pulse was applied, and the voltage drops were recorded. The ohmic resistance RΩ, polarization resistance Rpol, and total resistance R were calculated using the following equations:
$$
R_{\Omega} = \frac{\Delta V_1}{I} \tag{1}
$$
$$
R_{\text{pol}} = \frac{\Delta V_2 – \Delta V_1}{I} \tag{2}
$$
$$
R = R_{\Omega} + R_{\text{pol}} \tag{3}
$$
where I is the pulse current, ΔV1 is the instantaneous voltage drop at the start of the pulse, and ΔV2 is the voltage drop at the end of the 10 s pulse. The entropic heat coefficient dU/dT was determined by the open-circuit voltage method. The energy storage cell was equilibrated at a fixed SOC, then exposed to eight temperature setpoints (−15, 0, 15, 20, 25, 30, 35, 40 °C) in sequence, each held for 4 h. The stable open-circuit voltage was recorded, and the coefficient was computed as:
$$
\frac{dU}{dT} = \frac{\Delta U}{\Delta T} \tag{4}
$$
We repeated this at different SOC levels to obtain the full profile. All temperature measurements for this test used the center point T3 as reference.
2. Results and Discussion
2.1 Heat Generation Behavior of the Energy Storage Cell
The internal resistance of the large-capacity energy storage cell exhibited moderate dependence on SOC and discharge rate. As shown in Table 2, the total resistance remained around 0.5 mΩ for SOC between 10% and 90%, but increased at the extremes. At low SOC, higher discharge rates increased the resistance due to concentration polarization, while at high SOC the trend was reversed. The entropic heat coefficient was positive in the mid-SOC region (10%–90%), indicating endothermic reversible reactions that partially offset the irreversible Joule and polarization heat. At very low SOC (<10%) and very high SOC (>90%), the coefficient became negative, meaning the reversible reactions were exothermic, adding to the overall heat generation and making the energy storage cell thermally unstable at those conditions.
| SOC (%) | 0.25C | 0.5C | 1.0C |
|---|---|---|---|
| 10 | 0.48 | 0.51 | 0.55 |
| 30 | 0.47 | 0.49 | 0.50 |
| 50 | 0.46 | 0.47 | 0.48 |
| 70 | 0.47 | 0.46 | 0.45 |
| 90 | 0.49 | 0.48 | 0.47 |
We then investigated the thermal response of the energy storage cell during constant-current discharge at different C-rates (0.25C, 0.5C, 1.0C) in a 25 °C ambient environment, using point T3 (center) as the reference. The surface temperature rise and heat flux are shown in Table 3. A clear positive correlation between discharge rate and temperature rise was observed. At 1.0C, the maximum temperature reached 46.64 °C (ΔT = 21.64 °C), while at 0.25C it was only 28.5 °C (ΔT = 3.5 °C). The peak heat flux increased from a negligible level at 0.25C to over 600 W/m2 at 1.0C, indicating dramatically higher heat generation rates at high rates. Moreover, during 0.25C discharge we observed a slight temperature decrease of 1–2 °C in the early stage, because the endothermic reversible heat surpassed the Joule heat. At high rates, the irreversible heat dominated and the temperature rose monotonically.
| Discharge rate | Max. temperature at T3 (°C) | Temperature rise ΔT (°C) | Peak heat flux (W/m2) |
|---|---|---|---|
| 0.25C | 28.50 | 3.50 | ~50 |
| 0.5C | 35.20 | 10.20 | ~250 |
| 1.0C | 46.64 | 21.64 | ~600 |
Spatial temperature distribution across the large-format energy storage cell was highly non-uniform. Figure 1 illustrates the temperature rise at eight monitoring points during 1.0C discharge. The negative-terminal region (T1) consistently experienced the highest temperature rise, while the bottom area (T4) remained the coolest. This non-uniformity became more pronounced at higher rates. For example, during 1.0C discharge, the temperature difference between T1 and T4 reached approximately 8 °C. The top region of the cell (T1, T2, T6) generally showed more intense heat accumulation than the bottom. This is likely due to the fact that the negative electrode (graphite) generates more heat during delithiation, and the tab connections contribute additional resistive heating. The heat flux also varied spatially, with the highest values measured at the 1/3 height location near the negative tab.
During charging, we observed a similar pattern. The temperature rise rate peaked at 0.54 °C/min during the constant-current phase of 1.0C charge, while the subsequent constant-voltage phase generated less heat due to decreasing current. In contrast, during 1.0C discharge the temperature rise rate reached 0.7 °C/min, and the temperature continued to rise until the end of discharge because of increasing internal resistance at low SOC. These findings emphasize that high-rate operation, especially discharging, imposes severe thermal stress on the energy storage cell.
We further examined the combined effect of ambient temperature and discharge rate on the temperature rise of the energy storage cell. Table 4 summarizes the ΔT at different ambients and rates. At lower ambient temperatures (e.g., 15 °C), the temperature rise was larger than at 25 °C or 45 °C for the same rate because the increased electrolyte viscosity and reduced conductivity led to higher internal resistance and more heat generation. At elevated ambient temperatures (35 °C, 45 °C), the internal resistance decreased, and the temperature rise was slightly lower. However, the absolute temperature reached higher values, which is detrimental to long-term cycling stability.
| Ambient temperature (°C) | 0.25C discharge | 0.5C discharge | 1.0C discharge |
|---|---|---|---|
| 15 | 4.2 | 12.5 | 25.3 |
| 25 | 3.5 | 10.2 | 21.6 |
| 35 | 2.8 | 8.6 | 18.4 |
| 45 | 2.1 | 6.9 | 15.2 |
The heat flux measurements confirmed that thermal generation was higher at lower ambients, and the spatial non-uniformity persisted regardless of ambient temperature. The axial gradient showed that the 1/3 height region always generated the highest heat flux, especially during the initial stage of discharge. This indicates that local thermal management, such as enhanced cooling near the negative tab, is critical for large-capacity energy storage cells.
2.2 Capacity Degradation and Cycling Performance
To understand how thermal conditions affect the long-term aging of the energy storage cell, we conducted 100 full charge-discharge cycles at 1.0C rate under three ambient temperatures: 25 °C, 35 °C, and 45 °C. A 4-hour rest was inserted between charge and discharge. The capacity retention curves are plotted in Figure 2 (simulated data). Initially, the cells at 35 °C and 45 °C showed slightly higher capacity than at 25 °C due to enhanced ionic conductivity and reduced internal resistance. However, after only a few cycles, the high-temperature cells began to degrade faster. By the 100th cycle, the capacity fade was 1.61% at 25 °C, 2.45% at 35 °C, and 3.64% at 45 °C. The capacity loss in Ah was 4.09, 6.96, and 10.31 Ah, respectively. The degradation rate at 45 °C was 2.26 times that at 25 °C. This accelerated aging is attributed to several mechanisms: increased solid electrolyte interphase (SEI) growth, electrolyte decomposition, structural degradation of the LiFePO4 cathode, and lithium plating at the anode under high temperature.
| Ambient temperature (°C) | Capacity fade (Ah) | Capacity retention (%) | Normalized fade rate |
|---|---|---|---|
| 25 | 4.09 | 98.39 | 1.00 |
| 35 | 6.96 | 97.55 | 1.70 |
| 45 | 10.31 | 96.36 | 2.52 |
Internal resistance evolution during cycling also confirmed the detrimental effect of high temperature. Table 6 presents the direct current internal resistance (DCIR), ohmic resistance (RΩ), and polarization resistance (Rpol) measured at 50% SOC after every 20 cycles. At 25 °C, the resistances increased slowly. At 45 °C, the polarization resistance rose sharply, from 0.12 mΩ initially to 0.31 mΩ after 100 cycles, indicating severe electrode polarization and loss of active material. Ohmic resistance also increased, but at a slower pace.
| Cycle number | DCIR (mΩ) | RΩ (mΩ) | Rpol (mΩ) |
|---|---|---|---|
| 0 | 0.50 | 0.32 | 0.18 |
| 20 | 0.52 | 0.33 | 0.19 |
| 40 | 0.55 | 0.34 | 0.21 |
| 60 | 0.58 | 0.35 | 0.23 |
| 80 | 0.62 | 0.36 | 0.26 |
| 100 | 0.67 | 0.37 | 0.30 |
We further analyzed the incremental capacity (dQ/dV) curves to understand aging mechanisms. The dQ/dV profiles for the energy storage cell showed three characteristic peaks (Q1, Q2, Q3) corresponding to phase transitions in the graphite anode and LiFePO4 cathode. Over cycling, the peaks shifted and their areas decreased. Table 7 summarizes the peak areas after 0, 50, 100 cycles at different temperatures. The high-potential peak Q3 degraded fastest, especially at 45 °C, where its area dropped by 12% after 100 cycles, compared to only 4% at 25 °C. This suggests that the cathode material is more vulnerable at elevated temperatures, possibly due to iron dissolution and structural changes. The low-potential peaks Q1 and Q2 also decreased, but more moderately, indicating loss of active lithium and anode degradation.
| Temperature | Cycle | Q1 (norm.) | Q2 (norm.) | Q3 (norm.) |
|---|---|---|---|---|
| 25 °C | 0 | 1.000 | 1.000 | 1.000 |
| 50 | 0.986 | 0.985 | 0.979 | |
| 100 | 0.972 | 0.970 | 0.960 | |
| 35 °C | 0 | 1.000 | 1.000 | 1.000 |
| 50 | 0.978 | 0.975 | 0.965 | |
| 100 | 0.955 | 0.950 | 0.932 | |
| 45 °C | 0 | 1.000 | 1.000 | 1.000 |
| 50 | 0.965 | 0.960 | 0.945 | |
| 100 | 0.935 | 0.925 | 0.880 |
The consistent pattern across all tests is that high temperatures accelerate capacity fade in large-capacity energy storage cells. The degradation is closely linked to increased polarization resistance and loss of active material, particularly at the cathode. Our findings highlight the importance of maintaining the energy storage cell within an optimal temperature window, ideally around 25 °C, to balance performance and longevity.
3. Conclusions
Through systematic experimental investigation of a 280 Ah prismatic LiFePO4 energy storage cell, we have characterized its heat generation behavior and capacity degradation under different operating conditions. The following conclusions are drawn:
- Heat generation strongly depends on discharge rate and ambient temperature. At 1.0C discharge, the temperature rise reached 21.64 °C, six times that at 0.25C. Lower ambient temperatures further increased the temperature rise due to higher internal resistance. The spatial temperature distribution across the large-format energy storage cell was highly non-uniform, with the negative terminal region consistently being the hottest. This implies that thermal management systems must prioritize cooling of the tab areas to mitigate local hot spots.
- High temperature accelerates capacity fade. After 100 cycles at 1.0C, the energy storage cell lost 3.64% capacity at 45 °C, compared to only 1.61% at 25 °C. The degradation rate increased by a factor of 2.26 per 20 °C rise. The dQ/dV analysis revealed that the high-potential peak (Q3) degraded most rapidly at elevated temperatures, indicating cathode material instability as a key aging mechanism.
- Implications for battery management. To maximize the lifetime of large-capacity energy storage cells, we recommend maintaining the operating temperature below 35 °C and preferably near 25 °C. Dynamic thermal management strategies that adapt to real-time heat generation, such as variable coolant flow or phase-change materials, should focus on the top region of the cell where heat accumulates most. Additionally, limiting high-rate charge/discharge cycles, especially at low SOC, can reduce thermal stress and prolong cycle life.
Our work provides a quantitative foundation for the safe and efficient design of large-capacity energy storage systems. Future research should explore more advanced cooling approaches and investigate aging under real-world combined thermal and electrical stress profiles. The insights gained here will contribute to the development of reliable and durable energy storage solutions.