In the context of global carbon neutrality and peak carbon goals, renewable energy sources such as wind and solar power are progressively replacing fossil fuels in the power grid. However, their inherent intermittency and volatility necessitate robust energy storage systems to ensure grid stability. Among various electrochemical storage technologies, the lithium-ion energy storage battery has emerged as a leading candidate due to its high energy density, long cycle life, high efficiency, and declining cost. Yet, during high-power operations—such as fast charging and frequency regulation—the internal Ohmic resistance and electrochemical reactions generate substantial heat, with heat flux increasing threefold or more. Traditional thermal management methods, such as forced air cooling or single-sided liquid cooling, face critical limitations in meeting these demands. This work presents an innovative gas-liquid coupled dual-phase thermal management design based on a lithium iron phosphate (LFP) energy storage battery pack, addressing the extreme cooling requirements under 3C-rate charge-discharge cycles.
Our proposed system integrates forced air cooling at the top of the battery cells with an indirect liquid cooling plate at the bottom. After the gaseous medium (air) absorbs heat from the upper portion of the cells, it flows downward to a finned liquid cooling plate, where it exchanges heat with the coolant. The cooled air then recirculates to the top, creating a closed-loop dual-phase heat exchange cycle. This approach eliminates the condensation risks associated with top-mounted liquid cooling plates while significantly increasing the effective heat transfer area. To systematically evaluate this design, we conducted three-dimensional computational fluid dynamics (CFD) simulations using Flotherm-XT software, employing a 1P48S battery pack configuration (48 cells in series) under a 3C-rate charge-discharge cycle.
Simulation Model and Governing Parameters
The simulation model comprises a battery pack with dimensions of 900 mm × 670 mm × 225 mm. Eight miniature fans (60 × 60 × 15 mm, each with a power of 1.4 W) are arranged on both sides to drive forced air flow across the top surfaces of the cells. The liquid cooling plate uses a water-glycol mixture at a flow rate of 10 L/min and an inlet temperature of 20 °C. A structured mesh with adaptive refinement was used, and grid independence was verified at 2 million elements. The k-ε turbulence model was selected for convective heat transfer calculations. The key material properties are summarized in Table 1.
| Component | Density (kg/m³) | Specific Heat (J/(kg·K)) | Thermal Conductivity (W/(m·K)) | Dynamic Viscosity (Pa·s) |
|---|---|---|---|---|
| Battery cell (LFP) | 2205.5 | 1008.7 | X:17.8, Y:4.2, Z:2.3 | / |
| Coolant (water-glycol) | 1077 | 3473 | 0.4 | 0.0046 |
| Cold plate (aluminum) | 2700 | 900 | 177 | / |
| Thermal adhesive | 2000 | 1200 | 1.5 | / |
| Aluminum busbar | 2702 | 903 | 237 | / |
| Enclosure (steel) | 7850 | 470 | 52 | / |
| End plate (aluminum) | 2760 | 963 | 96.2 | / |
| Aerogel insulation | 320 | 1000 | 0.025 | / |
The battery heat generation profile during a 3C charge-discharge cycle is defined by a time-dependent function, derived from experimental data. The total heat generation for the pack is:
$$Q_{\text{total}} = 48 \times \int_{0}^{2400} P(t) \, dt$$
where the time step is 120 s and the trapezoidal rule is applied for numerical integration.
Results: Temperature Performance Comparison
Under the 3C charge-discharge condition, we compared the thermal performance of three configurations: (1) a conventional bottom-only liquid cooled pack, (2) a gas-liquid coupled pack without flow channel separation, and (3) the proposed gas-liquid coupled pack with alternating gas flow channels. The final temperature distributions are summarized in Table 2.
| Configuration | Maximum Cell Temperature (°C) | Maximum Temperature Difference Within Pack (°C) | Cell Top-to-Bottom Temperature Difference (°C) |
|---|---|---|---|
| Bottom-only liquid cooling | 54.0 | 15.0 | 20.0 |
| Gas-liquid coupled (basic, no channel separation) | 51.5 | 12.0 | 14.5 |
| Gas-liquid coupled (with alternating gas channels) | 46.5 | 8.5 | 14.0 |
| Gas-liquid coupled (optimized fin gap and corrugated fins) | 46.0 | 8.5 | 13.5 |
The conventional bottom-only liquid cooled pack exhibits a maximum temperature of 54.0 °C and an intra-pack temperature difference of 15.0 °C. In contrast, our optimized gas-liquid coupled design reduces these values to 46.0 °C and 8.5 °C, representing reductions of 8.0 °C and 6.5 °C, respectively. The reduction in the cell top-to-bottom temperature gradient from 20.0 °C to 13.5 °C is particularly important for mitigating capacity fade and internal resistance variation. The alternating gas flow channel design ensures that the gaseous medium maintains a lower temperature throughout its path by preventing premature warming of the air contacting downstream cells.
Theoretical Analysis: Fin Gap Optimization
To maximize the total heat transfer rate \(Q\) from the top air cooling system, we performed an analytical optimization of the fin gap in the alternating flow channels using the effectiveness-NTU (ε-NTU) method. The governing equations are:
$$Q = \varepsilon \dot{m} C_p (T_f – T_w)$$
$$\varepsilon = 1 – \exp(-\text{NTU})$$
$$\text{NTU} = \frac{h A}{\dot{m} C_p}$$
where \(T_f\) is the inlet air temperature, \(T_w\) is the cold plate wall temperature, \(\dot{m}\) is the mass flow rate, and \(A\) is the total heat transfer area. The convective heat transfer coefficient \(h\) for laminar flow scales as:
$$h \propto u^{0.5} d^{-0.5}$$
where \(u\) is the air velocity and \(d\) is the fin gap. Under constant fan power \(P\):
$$P = \Delta P \cdot \dot{V}$$
The pressure drop \(\Delta P\) for laminar flow in a rectangular channel scales as \(\Delta P \propto \mu u L / d^2\). Combining these relationships:
$$u \propto d$$
Substituting into the NTU expression:
$$\text{NTU} \propto u^{-0.5} d^{-1.5} (2W + d)$$
where \(W\) is the fin height. The total heat transfer rate becomes:
$$Q \propto [1 – \exp(-\text{NTU})] \cdot \dot{m}$$
$$\dot{m} \propto \frac{d^2}{t + d}$$
where \(t\) is the fin thickness. The function exhibits a maximum at an optimal fin gap, which we determined through parametric CFD simulations. The results are presented in Table 3.
| Fin Gap (mm) | Average Top Busbar Temperature (°C) | Pack Maximum Temperature (°C) | Pack Temperature Difference (°C) |
|---|---|---|---|
| 3 | 45.5 | 49.0 | 10.0 |
| 5 | 46.8 | 50.2 | 9.5 |
| 8 | 47.2 | 50.8 | 8.8 |
| 11 | 48.0 | 51.5 | 6.0 |
| 14 | 49.5 | 52.0 | 8.0 |
| 18 | 50.0 | 53.0 | 9.2 |
The optimal fin gap was found to be 11 mm, which minimizes the intra-pack temperature difference to 6.0 °C while keeping the maximum temperature within acceptable limits. However, the best overall trade-off between maximum temperature and uniformity was achieved at 3 mm, which gave the lowest average busbar temperature of 45.5 °C.
Bottom Cold Plate Enhancement: Corrugated Fins
To further improve the cooling capacity of the bottom liquid cold plate and lower the temperature of the recirculating air, we replaced the straight fins with corrugated (sinusoidal wave) fins. The corrugation induces flow separation and secondary flows, enhancing the convective heat transfer coefficient. The fin pitch and wave period were systematically optimized. The heat transfer rate from the air to the cold plate is governed by:
$$Q_{\text{bottom}} = h_{\text{eff}} A_{\text{eff}} \Delta T_{\text{lm}}$$
where \(h_{\text{eff}}\) is the effective heat transfer coefficient, \(A_{\text{eff}}\) is the effective wetted area (increased by 45% compared to straight fins), and \(\Delta T_{\text{lm}}\) is the log-mean temperature difference. We examined wave periods ranging from 50 mm to 130 mm. Table 4 summarizes the simulation results.
| Wave Period (mm) | Pack Maximum Temperature (°C) | Pack Temperature Difference (°C) |
|---|---|---|
| 50 | 49.0 | 10.2 |
| 70 | 47.5 | 9.0 |
| 90 | 46.0 | 8.5 |
| 110 | 47.0 | 9.5 |
| 130 | 49.0 | 10.0 |
The optimal wave period of 90 mm yields the lowest maximum temperature (46.0 °C) and the smallest pack temperature difference (8.5 °C). The corrugated fins improve heat transfer through four mechanisms: (1) disruption of the laminar boundary layer, (2) increased effective surface area (by up to 45%), (3) extended flow path length for better thermal equilibration, and (4) generation of secondary flows that homogenize temperature profiles. The fin pitch for the corrugated design was optimized at 6 mm, compared to 3 mm for the straight fins, due to the altered flow dynamics.
Heat Transfer Efficiency Calculation
We quantified the heat transfer efficiency of each configuration using the total heat absorbed by the coolant. The energy stored in the battery pack is:
$$Q_{\text{stored}} = N_{\text{cells}} \cdot C_{\text{cell}} \cdot M_{\text{cell}} \cdot \Delta T_{\text{avg}}$$
where \(N_{\text{cells}} = 48\), \(C_{\text{cell}} = 1008.7 \, \text{J/(kg·K)}\), \(M_{\text{cell}}\) is the cell mass, and \(\Delta T_{\text{avg}}\) is the average temperature rise. The total heat generated during the 3C cycle is \(Q_{\text{gen}} = 3,755,904 \, \text{J}\). The heat removed by the cooling system is \(Q_{\text{removed}} = Q_{\text{gen}} – Q_{\text{stored}}\). Table 5 compares the efficiency of the two thermal management strategies.
| Configuration | Heat Stored in Cells (J) | Heat Removed by Cooling (J) | Heat Transfer Efficiency (%) |
|---|---|---|---|
| Bottom-only liquid cooling | 1,464,802 | 2,291,102 | 61.0 |
| Optimized gas-liquid coupled | 938,976 | 2,816,928 | 75.0 |
The gas-liquid coupled design improves heat transfer efficiency from 61% to 75%, a relative increase of 23%. This enhancement is attributed to the dual-path heat extraction: the top air cooling handles approximately 30% of the total heat load, while the bottom liquid cooling handles the remaining 70%, with the two paths coupled through the recirculating air.
Lifetime Prediction: Impact of Temperature Reduction
The temperature reduction achieved by the gas-liquid coupled design has a direct and significant impact on the cycle life of the LFP energy storage battery. Experimental data from our laboratory indicates that for LFP cells operating above 30 °C, every 5 °C increase in average temperature accelerates capacity fade by approximately 20%. Figure 8 in the original study shows that increasing the operating temperature from 45 °C to 50 °C reduces cycle life by over 1,000 cycles. Given our optimized design reduces the average pack temperature from 54 °C to 46 °C—an 8 °C reduction—the projected increase in cycle life is substantial.
The capacity retention over cycles can be approximated by an Arrhenius-type degradation model:
$$\frac{dQ}{dN} = -k(T) \cdot Q$$
$$k(T) = k_0 \exp\left(-\frac{E_a}{RT}\right)$$
where \(Q\) is the remaining capacity, \(N\) is the cycle number, \(k_0\) is a pre-exponential factor, and \(E_a\) is the activation energy for degradation. For LFP cells, \(E_a \approx 30 \, \text{kJ/mol}\). The relative lifetime improvement is calculated as:
$$\frac{L_{46}}{L_{54}} = \exp\left(\frac{E_a}{R} \left(\frac{1}{T_{46}} – \frac{1}{T_{54}}\right)\right)$$
Using \(T_{46} = 319.15 \, \text{K}\) and \(T_{54} = 327.15 \, \text{K}\), with \(R = 8.314 \, \text{J/(mol·K)}\) and \(E_a = 30,000 \, \text{J/mol}\):
$$\frac{L_{46}}{L_{54}} \approx \exp\left(\frac{30000}{8.314} \left(\frac{1}{319.15} – \frac{1}{327.15}\right)\right) \approx 1.23$$
This corresponds to a 23% increase in cycle life. For a baseline life of 6,000 cycles at 54 °C, the optimized gas-liquid coupled energy storage battery pack achieves approximately 7,400 cycles, a gain of 1,400 cycles. This improvement translates directly to lower total cost of ownership and reduced maintenance frequency for grid-scale energy storage systems.
Conclusion and Outlook
This work presents a novel gas-liquid coupled dual-phase thermal management system for high-power LFP energy storage battery packs. The design leverages forced air cooling at the top of the cells coupled with a fin-enhanced indirect liquid cooling plate at the bottom, creating a closed-loop heat exchange cycle. Through systematic optimization of the alternating gas flow channels, fin gaps, and corrugated fin geometry, we achieved a maximum cell temperature of 46.0 °C and a pack temperature difference of 8.5 °C under 3C charge-discharge conditions—improvements of 8.0 °C and 6.5 °C, respectively, compared to conventional bottom-only liquid cooling. The heat transfer efficiency increased from 61% to 75%, and the projected cycle life improved by 1,400 cycles, representing a 23% enhancement.
The proposed thermal management concept offers a compelling solution for high-power battery systems in electric vehicles, frequency regulation, and fast-charging applications. The elimination of condensation risks, combined with the ability to retrofit existing bottom-cooled packs with simple air channel modifications, makes this approach highly practical. Future work will focus on validating the design under extreme environmental conditions, including low-temperature startup (-20 °C) and high-temperature operation (45 °C), as well as characterizing performance under pulsed high-rate (5C) discharges. Additionally, the integration of the gas-liquid coupled design into building-integrated energy storage and data center thermal management presents promising opportunities for cross-sector applications.