In the realm of modern energy solutions, the battery energy storage system has become a cornerstone for applications ranging from electric vehicles to grid-scale storage. As a researcher deeply involved in thermal management technologies, I have focused on addressing the critical challenges associated with lithium-ion batteries, which are prone to temperature rise during charge-discharge cycles. Excessive heat not only degrades efficiency and accelerates aging but can also lead to thermal runaway, posing significant safety risks. Therefore, effective thermal management is paramount for optimizing the performance and longevity of any battery energy storage system.
Traditional cooling methods, such as air cooling and phase-change material cooling, often fall short under extreme conditions. In recent years, liquid cooling has emerged as a more efficient alternative, with immersion cooling—where the coolant directly contacts the battery cells—gaining traction for its superior temperature uniformity and heat dissipation capabilities. However, current immersion cooling systems predominantly rely on fluorinated liquids, which suffer from high costs, poor thermophysical properties, and environmental concerns. Alternative media like silicone oils offer better thermal performance and cost-effectiveness but introduce flammability risks due to their flash points. This study aims to overcome these limitations by proposing a novel dual-circuit dynamic switching system for immersion cooling in battery energy storage systems, utilizing modified silicone oil as the primary coolant and integrating a fire-suppression mechanism for enhanced safety.
The selection of an appropriate coolant is pivotal for the efficiency and safety of an immersion-cooled battery energy storage system. Based on extensive literature review and industry data, I compared two prominent candidates: AC6000 fluorinated liquid and ICL-1000 modified silicone oil. The key performance parameters are summarized in the table below, highlighting the advantages of modified silicone oil in terms of thermal conductivity, specific heat capacity, and dielectric strength, albeit with higher viscosity that necessitates optimized flow design.
| Performance Indicator | AC6000 Fluorinated Liquid | ICL-1000 Modified Silicone Oil | Comparative Analysis |
|---|---|---|---|
| Flash Point (°C) | Non-flammable | ≥320 | Fluorinated liquid has no flash point advantage, but silicone oil achieves ultra-high flash point through modification, significantly reducing flammability risks. |
| Thermal Conductivity (W/m·K) | 0.06–0.07 | 0.15–0.18 | Silicone oil’s thermal conductivity is 2.5 times higher, facilitating faster heat dissipation in the battery energy storage system. |
| Specific Heat Capacity (kJ/kg·K) | 1.05–1.10 | 1.65–1.75 | Silicone oil exhibits 60% higher heat storage capacity, which slows the temperature rise rate during operation. |
| Viscosity @40°C (10⁻⁶ m²/s) | 0.6–0.8 | 18–22 | Fluorinated liquid’s low viscosity reduces pump power requirements, whereas silicone oil’s higher viscosity demands careful flow channel optimization. |
| Dielectric Strength (kV/mm) | ≥35 | ≥45 | Silicone oil offers superior insulation properties, minimizing short-circuit risks in the battery pack. |
Beyond performance, cost considerations are crucial for scaling immersion cooling in battery energy storage systems. The following table illustrates the stark cost difference between the two media, with modified silicone oil being substantially more economical.
| Cost Item | AC6000 Scheme | ICL-1000 Scheme |
|---|---|---|
| Medium Cost (10k CNY/MW·h) | 9.8 | 2.7 |
From a thermal perspective, the heat transfer efficiency can be quantified using the convective heat transfer equation: $$Q = hA\Delta T$$ where \(Q\) is the heat flux, \(h\) is the convective heat transfer coefficient, \(A\) is the surface area, and \(\Delta T\) is the temperature difference. For immersion cooling, \(h\) is influenced by the coolant’s properties, such as thermal conductivity \(k\) and viscosity \(\mu\). The Nusselt number \(Nu\), which relates convective to conductive heat transfer, can be expressed as $$Nu = \frac{hD}{k}$$ where \(D\) is a characteristic length. For laminar flow in channels, \(Nu\) often depends on the Reynolds number \(Re\) and Prandtl number \(Pr\), given by $$Re = \frac{\rho v D}{\mu} \quad \text{and} \quad Pr = \frac{\mu c_p}{k}$$ with \(\rho\) as density, \(v\) as velocity, and \(c_p\) as specific heat capacity. Modified silicone oil’s higher \(k\) and \(c_p\) enhance \(Nu\), leading to better cooling performance in a battery energy storage system.
However, the flash point of silicone oil—though elevated through modification—still poses a risk at extreme temperatures above 250°C. To address this, I designed a dual-circuit dynamic switching system that integrates cooling and fire suppression within a single battery pack enclosure. The core innovation lies in using shared flow channels to switch between modified silicone oil for routine cooling and a fire-suppressant liquid (e.g., perfluorohexanone) during thermal runaway events. This approach ensures both efficient thermal management and safety without the high costs associated with fluorinated liquids.
The battery pack structure comprises an installation chassis, isolation frames, separation plates, and sealing strips. The chassis features a cooling trench along its length, with inlet and outlet pipes for coolant circulation. Battery cells are mounted on isolation plates, and the entire assembly is immersed in coolant. The dual-circuit system operates based on temperature thresholds: under normal conditions, modified silicone oil circulates to absorb heat; if temperature sensors detect a cell temperature exceeding 250°C, controllers activate valves to disconnect the coolant loop and inject fire-suppressant liquid, rapidly displacing the silicone oil to quench any incipient fire. Once temperatures drop below 80°C, the system switches back to silicone oil, and the suppressant is filtered for reuse. This design not only improves safety but also optimizes cost by sharing components between circuits.
To validate the thermal and fluid dynamic performance of this system, I conducted comprehensive simulations using computational fluid dynamics (CFD) software. The internal flow channel model was constructed with parameters aligned to the modified silicone oil ICL-1000. The governing equations for fluid flow and heat transfer include the Navier-Stokes equations for momentum conservation: $$\rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla p + \mu \nabla^2 \mathbf{v} + \mathbf{f}$$ and the energy equation for temperature distribution: $$\rho c_p \left( \frac{\partial T}{\partial t} + \mathbf{v} \cdot \nabla T \right) = k \nabla^2 T + \dot{q}$$ where \(\mathbf{v}\) is the velocity vector, \(p\) is pressure, \(\mathbf{f}\) represents body forces, and \(\dot{q}\) is the heat generation rate per unit volume from battery cells. For the battery energy storage system, \(\dot{q}\) varies with charge-discharge cycles, modeled based on operational data.
The flow field simulation focused on pressure drop and velocity distribution to ensure efficient coolant circulation. The boundary conditions and medium properties are listed below.
| Boundary Condition | Parameter |
|---|---|
| Inlet Flow Rate (L/min) | 4 |
| Reynolds Number (Re) | 583 |
| Medium Name | ICL-1000 Modified Silicone Oil |
|---|---|
| Density (kg/m³) | 860 |
| Viscosity @40°C (10⁻⁶ m²/s) | 22 |
| Specific Heat Capacity (kJ/kg·K) | 1.75 |
| Thermal Conductivity (W/m·K) | 0.18 |
The simulation results showed an inlet-outlet pressure drop of 3.8 kPa, with a velocity contour indicating minimal flow stagnation zones. The Reynolds number, calculated as $$Re = \frac{\rho v D_h}{\mu}$$ where \(D_h\) is the hydraulic diameter, confirmed laminar flow conditions, which are typical for immersion cooling in compact battery energy storage systems. The velocity profile remained uniform across the flow channels, ensuring consistent cooling contact with all battery cells.
For the thermal field simulation, I developed a model incorporating various materials used in the battery pack, such as aluminum alloys for structural components and thermal conductive adhesives for interface layers. The material properties are summarized in the following table, which are critical for accurate heat transfer analysis in the battery energy storage system.
| Structure | Material | Density (kg/m³) | Thermal Conductivity (W/m·K) | Specific Heat Capacity (J/kg·K) | Viscosity @40°C (cSt) |
|---|---|---|---|---|---|
| Battery Cell | — | 2161.9 | Normal: 5.13, Transverse: 23.77 | 1000 | — |
| Foam Rubber | — | 60 | 0.034 | 1000 | — |
| Epoxy Board | — | 1800 | 0.2 | 550 | — |
| End Plate | AL6063 | 2702 | 218 | 900 | — |
| Cross Beam | AL6063 | 2702 | 218 | 900 | — |
| Reinforcing Rib | AL6063 | 2702 | 218 | 900 | — |
| Thermal Conductive Adhesive | — | 2500 | 1.5 | 1000 | — |
| Liquid Cooling Plate | AL3003 | 2702 | 155 | 893 | — |
| Coolant | Modified Silicone Oil | 860 | 0.18 | 1750 | 22 |
The boundary conditions for the thermal simulation included ambient temperature, battery operational cycles, and cooling parameters, as detailed below.
| Boundary Condition | Parameter |
|---|---|
| Environment and Initial Temperature | 25°C |
| Cell Operating Condition | 0.5C charge for 2h + 0.5C discharge for 2h Charge Heat Generation Power: 11.542 W Discharge Heat Generation Power: 10.964 W |
| Coolant Flow Rate | 4 L/min, 3 L/min (for sensitivity analysis) |
| External Refrigeration Power | 500 W (minimum inlet coolant temperature 20°C) |
The heat generation from battery cells was modeled using an empirical formula based on the charge-discharge rates: $$\dot{q} = I^2 R + I \left( \frac{\partial U}{\partial T} \right) \Delta T$$ where \(I\) is the current, \(R\) is the internal resistance, and \(U\) is the open-circuit voltage. For the 0.5C cycle, the heat generation rates were derived from experimental data. The simulation solved the coupled equations iteratively, with convergence criteria set to residuals below \(10^{-6}\) for energy and \(10^{-5}\) for momentum.
The thermal simulation results revealed excellent temperature uniformity across the battery pack. The maximum temperature difference observed was only 1.814°C, which complies with national standards for electric vehicle battery packs and underscores the effectiveness of immersion cooling for battery energy storage systems. The temperature distribution can be described by the Fourier heat conduction law in differential form: $$ \nabla \cdot (k \nabla T) + \dot{q} = \rho c_p \frac{\partial T}{\partial t} $$ With the modified silicone oil’s high \(k\) and \(c_p\), the system achieves rapid heat dissipation, maintaining cells within the optimal 20–40°C range. The pressure drop, after accounting for adhesive layer thickness, was 3.2 kPa, indicating low pumping power requirements.
To further analyze the cooling performance, I calculated the overall heat transfer coefficient \(U\) for the immersion system using the formula: $$\frac{1}{U} = \frac{1}{h_i} + \frac{\delta}{k_w} + \frac{1}{h_o}$$ where \(h_i\) and \(h_o\) are the convective heat transfer coefficients on the cell and coolant sides, respectively, and \(\delta\) and \(k_w\) are the thickness and thermal conductivity of the wall materials. For modified silicone oil, \(h_o\) is enhanced due to its higher thermal conductivity, leading to a \(U\) value approximately 30% greater than that with fluorinated liquid, as per the relation $$h_o \propto \frac{k^{0.6} c_p^{0.4} \rho^{0.8} v^{0.8}}{\mu^{0.4} D^{0.2}}$$ derived from dimensional analysis for forced convection.
The dual-circuit switching mechanism adds a layer of safety without compromising cooling efficiency. The transition time from silicone oil to fire-suppressant liquid is critical and can be estimated using fluid displacement equations: $$t_{\text{switch}} = \frac{V_{\text{cavity}}}{Q_{\text{suppressant}}}$$ where \(V_{\text{cavity}}\) is the volume of the battery pack cavity and \(Q_{\text{suppressant}}\) is the flow rate of the suppressant. For a typical battery energy storage system with \(V_{\text{cavity}} = 0.1 \, \text{m}^3\) and \(Q_{\text{suppressant}} = 10 \, \text{L/s}\), \(t_{\text{switch}} \approx 10 \, \text{s}\), sufficiently fast to prevent thermal runaway propagation. The fire-suppressant, such as perfluorohexanone, works by both cooling and inerting, with its effectiveness quantified by the heat absorption capacity: $$Q_{\text{absorb}} = m c_p \Delta T + m L_v$$ where \(m\) is mass, \(c_p\) is specific heat, \(\Delta T\) is temperature change, and \(L_v\) is latent heat of vaporization if phase change occurs.
In terms of cost-benefit analysis, the dual-circuit system offers significant savings over traditional fluorinated liquid setups. The total cost \(C_{\text{total}}\) for a battery energy storage system can be expressed as $$C_{\text{total}} = C_{\text{medium}} + C_{\text{hardware}} + C_{\text{operational}}$$ where \(C_{\text{medium}}\) is the coolant cost, \(C_{\text{hardware}}\) includes pumps and valves, and \(C_{\text{operational}}\) covers pumping power. With modified silicone oil, \(C_{\text{medium}}\) is reduced by over 70%, as shown earlier. Moreover, the shared flow channels lower \(C_{\text{hardware}}\) by 15–20% compared to separate cooling and fire suppression systems. The operational cost savings arise from the lower viscosity of silicone oil relative to some alternatives, reducing pump power \(P_{\text{pump}}\) calculated as $$P_{\text{pump}} = \Delta p \cdot Q / \eta$$ where \(\Delta p\) is pressure drop, \(Q\) is flow rate, and \(\eta\) is pump efficiency.
Scalability of this immersion cooling approach for large-scale battery energy storage systems is promising. The modular design allows for stacking multiple battery packs with centralized cooling units. The temperature uniformity across modules can be maintained by optimizing the flow distribution network using Bernoulli’s principle: $$p + \frac{1}{2} \rho v^2 + \rho g h = \text{constant}$$ where \(p\) is pressure, \(v\) is velocity, \(g\) is gravity, and \(h\) is height. By designing symmetric inlet and outlet manifolds, flow imbalances are minimized, ensuring each cell within the battery energy storage system receives adequate cooling.
Environmental impact is another consideration. Modified silicone oils are generally less persistent in ecosystems compared to fluorinated liquids, which have high global warming potential and bioaccumulation tendencies. The fire-suppressant liquids used in the dual-circuit system, like perfluorohexanone, also have lower ozone depletion potential and shorter atmospheric lifetimes. Thus, this design aligns with sustainability goals for green battery energy storage systems.
Future work could involve experimental validation under real-world conditions, including cycle testing and thermal abuse scenarios. Additionally, integrating smart controls with machine learning algorithms could optimize the switching thresholds based on predictive analytics, further enhancing the safety and efficiency of battery energy storage systems. The use of advanced materials, such as graphene-enhanced coolants, might push the thermal performance boundaries even further.
In conclusion, this research presents a viable solution for immersion cooling in battery energy storage systems by combining modified silicone oil with a dual-circuit dynamic switching mechanism. The system addresses the cost, performance, and safety limitations of existing coolants, offering excellent temperature uniformity (ΔT ≤ 1.8°C), reduced costs, and enhanced fire suppression capabilities. Through rigorous simulation and analysis, I have demonstrated that this approach can significantly improve the thermal management of lithium-ion batteries, paving the way for wider adoption of immersion cooling in next-generation battery energy storage systems. The integration of efficient cooling and safety features ensures reliability and longevity, making it a compelling choice for applications ranging from electric vehicles to grid storage, ultimately contributing to a more sustainable energy future.