As the core component of electric vehicles (EVs) and large-scale energy storage systems, the lithium-ion battery is prized for its high energy density, low self-discharge rate, and long cycle life. However, its performance, longevity, and, most critically, its safety are intensely sensitive to temperature. The optimal operating temperature window for a li ion battery is typically between 15°C and 50°C, with a maximum permissible temperature difference within a module of less than 5°C to ensure uniform aging and performance. Exceeding these limits, especially under high charging/discharging rates or abuse conditions, accelerates degradation and can trigger thermal runaway—a dangerous, self-sustaining chain of exothermic reactions leading to fire or explosion. Therefore, effective thermal management is not merely an enhancement but a fundamental safety requirement for li ion battery systems. Among various cooling strategies, immersion cooling has emerged as a highly promising technology due to its superior heat transfer capability and inherent safety advantages.
Immersion cooling involves directly submerging the li ion battery or entire module into a dielectric fluid, or coolant. Heat generated during operation is absorbed by the coolant via sensible heat (temperature rise) or latent heat (phase change) and then transported away from the battery pack through a circulation system. This direct contact eliminates interfacial thermal resistance present in indirect cooling methods (like cold plates), significantly boosting heat transfer efficiency. Key advantages of immersion cooling for li ion battery include:
- High Heat Transfer Coefficient: Liquids offer orders of magnitude higher convective heat transfer coefficients than air or passive phase change materials (PCMs).
- Excellent Temperature Uniformity: The direct and often turbulent contact promotes even heat distribution, minimizing hot spots within a li ion battery pack.
- Inherent Safety Potential: Many suitable coolants are electrically insulating and non-flammable, acting as a built-in fire suppressant that can delay or prevent thermal runaway propagation.
- System Simplicity: It simplifies pack design by removing the need for complex thermal interface materials and intricate cold plate channels.
The core working mechanism relies on convective heat transfer. The governing equation for heat removal by the coolant is given by Newton’s law of cooling:
$$ q = h A (T_{battery} – T_{coolant}) $$
where \(q\) is the heat transfer rate, \(h\) is the convective heat transfer coefficient, \(A\) is the wetted surface area of the li ion battery, \(T_{battery}\) is the battery surface temperature, and \(T_{coolant}\) is the bulk coolant temperature. Immersion cooling maximizes both \(A\) (full surface contact) and \(h\) (through fluid properties and flow dynamics), thereby maximizing \(q\) for efficient cooling.
Dielectric Coolants for Lithium-Ion Battery Immersion
The selection of coolant is paramount, dictating the system’s performance, stability, and safety. An ideal coolant for li ion battery immersion must possess high dielectric strength to prevent short circuits, chemical stability and compatibility with battery materials, high thermal capacity, low viscosity to reduce pumping power, and a high flash/auto-ignition point for safety. The primary categories are detailed below.
Traditional Coolants
Mineral Oils: These hydrocarbon-based fluids are widely used due to their high specific heat, excellent dielectric properties, high flash points, and chemical inertness. They are non-corrosive and non-toxic, providing a stable long-term environment for the li ion battery.
$$ \rho \approx 924 \, \text{kg/m}^3, \quad c_p \approx 1900 \, \text{J/(kg·K)}, \quad k \approx 0.13 \, \text{W/(m·K)} $$
Silicone Oils: Known for exceptional chemical and thermal stability across a wide temperature range, silicone oils are also good electrical insulators. Their viscosity can be tailored by varying polymer chain length, allowing optimization for different flow conditions.
$$ \rho \approx 970 \, \text{kg/m}^3, \quad c_p \approx 1370 \, \text{J/(kg·K)}, \quad k \approx 0.15 \, \text{W/(m·K)} $$
Fluorinated Liquids (e.g., HFE/Novec): Engineered fluids like 3M’s Novec series feature very low surface tension, low viscosity, zero ozone depletion potential, and non-flammability (for high fluorine-to-hydrogen ratio variants). Their boiling points can be tailored; for instance, Novec 7000 boils at 34°C, enabling efficient two-phase cooling within the optimal li ion battery temperature window.
$$ \rho \approx 1400 \, \text{kg/m}^3, \quad c_p \approx 1300 \, \text{J/(kg·K)}, \quad k \approx 0.075 \, \text{W/(m·K)} $$
Other fluids like natural/synthetic esters (higher biodegradability) and water-glycol mixtures (high \(c_p\) but high dielectric constant requiring strict sealing) are less common. The properties of common coolants are summarized in Table 1.
| Coolant | Dielectric Constant | Specific Heat, \(c_p\) (J/(kg·K)) | Thermal Conductivity, \(k\) (W/(m·K)) | Flash Point (°C) | Viscosity @20°C (mm²/s) | Density @20°C (kg/m³) |
|---|---|---|---|---|---|---|
| Mineral Oil | ~2.1 | ~1900 | ~0.130 | >160 | ~56.0 | ~924 |
| Silicone Oil | ~2.75 | ~1370 | ~0.150 | >140 | Varies widely | ~970 |
| HFE-7000 | ~7.4 | ~1300 | ~0.075 | None (Non-flammable) | ~0.32 | ~1400 |
| Water-Glycol | >50 | ~3470 | ~0.400 | >100 | ~4.9 | ~1080 |
Coolant Modification and Nanofluids
To overcome limitations like relatively low thermal conductivity of base oils, research focuses on formulating nanofluids—coolants with suspended nanoparticles (e.g., Al₂O₃, TiO₂, Cu, carbon nanotubes, graphene). The goals are to enhance thermal conductivity (\(k\)), modify viscosity (\(\mu\)), and potentially improve dielectric breakdown strength. The effective property can be modeled, though real behavior often deviates due to clustering and interfacial effects. A simplified model for thermal conductivity is:
$$ k_{nf} = k_{bf} \left(1 + \phi \cdot \frac{3(k_p/k_{bf} – 1)}{(k_p/k_{bf} + 2) – \phi(k_p/k_{bf} – 1)}\right) $$
where \(k_{nf}\), \(k_{bf}\), and \(k_p\) are the thermal conductivities of the nanofluid, base fluid, and particle, respectively, and \(\phi\) is the particle volume fraction. However, increasing \(\phi\) often leads to unacceptable increases in viscosity and settlement issues. Key research directions in nanofluid development for li ion battery cooling are highlighted in Table 2.
| Base Fluid | Nanoparticle | Concentration Range | Key Parameters Studied |
|---|---|---|---|
| Mineral Oil / Polyester | Al₂O₃ | 0.02–0.10 wt% | Thermal Conductivity, Viscosity |
| Mineral Oil / Polyester | MWCNT | 0.05–0.10 wt% | Thermal Conductivity, Viscosity |
| Mineral Oil | TiO₂, Cu, Ag | 0.00125–0.05 vol% | Thermal Conductivity, Viscosity |
| Silicone Oil | Graphene Nano-platelets | 0.05–0.15 wt% | Thermal Conductivity, Viscosity |
| Mineral Oil | Co₃O₄ | 0.0002–0.006 wt% | Breakdown Voltage |
| Mineral Oil | TiO₂ Nanosheets | 0.0025–0.01 wt% | Breakdown Voltage |
Cooling Performance and System Enhancement
Immersion cooling systems are broadly classified into single-phase and two-phase (boiling) systems, each with distinct mechanisms and research focuses.
Single-Phase Immersion Cooling
In single-phase systems, the coolant absorbs heat without changing phase. Heat transfer is governed by forced convection. The key parameter is flow velocity, which influences the Reynolds number (\(Re\)) and thus the flow regime and Nusselt number (\(Nu\)):
$$ Re = \frac{\rho v D_h}{\mu}, \quad Nu = \frac{h D_h}{k} $$
where \(v\) is velocity and \(D_h\) is the hydraulic diameter. Higher \(v\) increases \(h\), reducing the maximum temperature of the li ion battery. However, it also increases pumping power (\(P_{pump} \propto \Delta p \cdot \dot{V}\)) and can exacerbate temperature gradients if not properly managed. Studies have shown that single-phase immersion with oils can maintain li ion battery temperature below 40°C at moderate discharge rates with careful control of flow parameters like immersion depth and velocity. Table 3 summarizes key research findings.
| Coolant | Method | Key Parameters Explored | Reported Temp. Range | Module ΔT |
|---|---|---|---|---|
| Mineral/Therminol Oil | Experimental | Discharge Rate (C-rate) | 28–43°C | 2–11°C |
| Mineral Oil | Experimental | Flow Velocity | 19–40°C | <2.5°C |
| Transformer Oil | Experimental | Immersion Depth | 38–58°C | <1.5°C |
| Mineral Oil, Water, EF | Simulation | Thermal Performance, Pumping Power | 29–50°C | 0.1–5.0°C |
Two-Phase (Boiling) Immersion Cooling
This method utilizes coolants with boiling points within the operational range (e.g., HFE-7000, bp=34°C). When the li ion battery surface temperature exceeds the coolant’s saturation temperature, nucleate boiling occurs, absorbing large amounts of latent heat (\(q = \dot{m} h_{fg}\)). This results in extremely high effective heat transfer coefficients, excellent temperature homogeneity (as the boiling point acts as a temperature cap), but requires a condenser to complete the cycle. The heat transfer during nucleate boiling is complex but can be represented by correlations like the Rohsenow correlation:
$$ \frac{c_{p,l}(T_{wall} – T_{sat})}{h_{fg}} = C_{sf} \left[ \frac{q}{\mu_l h_{fg}} \sqrt{\frac{\sigma}{g(\rho_l – \rho_v)}} \right]^r Pr_l^s $$
where subscript \(l\) and \(v\) denote liquid and vapor, \(\sigma\) is surface tension, \(C_{sf}\) is an empirical constant, and \(r, s\) are exponents. Research demonstrates that two-phase immersion can keep a li ion battery below 40°C even at very high discharge rates (e.g., 20C) with minimal temperature spread (<2°C). The condensation subsystem, however, adds complexity and cost.
| Coolant | Method | Key Parameters Explored | Reported Temp. Range | Module ΔT |
|---|---|---|---|---|
| HFE-7000 | Experimental | Discharge Rate (Up to 20C) | 32–38°C | <1°C |
| HFE-7000 | Experimental | Intermittent Flow | <36°C | <2°C |
| HFE-7000 | Simulation/Exp. | Heat Transfer Coefficient | 31–42°C | 0.03–5.2°C |
System Structure Optimization
To address challenges like high system mass (due to heavy coolant) and high pumping power for viscous oils, structural optimizations are crucial. Research focuses on innovative flow channel designs and integration strategies:
- Manifold and Baffled Designs: Implementing manifold inlets/outlets or internal baffles can ensure uniform coolant distribution across all li ion battery cells, preventing localized overheating and reducing the maximum temperature spread. Computational studies show reductions in standard deviation of temperature by over 25% with optimized manifold designs.
- Multi-Pass and Cross-Flow Channels: Designing the flow path in multiple layers or cross-flow configurations enhances heat removal efficiency. Simulations indicate such designs can improve energy density by over 20% while drastically reducing pumping power consumption.
- Hybrid and Modular Cooling: Combining immersion with other techniques (e.g., tab cooling with immersed cells) or creating modular immersion units can target heat generation zones more precisely, reduce the total coolant volume (and weight), and improve overall system efficiency.
Mechanisms of Thermal Runaway Suppression
Beyond routine cooling, a critical advantage of immersion for li ion battery systems is its potential to mitigate thermal runaway (TR). The suppression operates on multiple levels:
- Prevention of Heat Accumulation: The high heat transfer coefficient continuously removes heat during cell failure, delaying the time to reach critical TR trigger temperatures (e.g., for SEI decomposition, anode-electrolyte reaction).
- Inerting and Flame Suppression: The cell is isolated from atmospheric oxygen. For non-flammable coolants (like fluorinated liquids), any ejected flammable gases from a failing li ion battery are diluted in an oxygen-deficient environment, quenching flame initiation or propagation.
- Chemical Inhibition: Emerging research suggests certain coolants, particularly fluorinated fluids, may participate in radical scavenging reactions at high temperatures. During TR, as temperatures exceed 500°C, these fluids can thermally decompose, releasing fluorine-based radicals (e.g., F•). These radicals can interrupt the chain-branching combustion reactions of hydrogen and hydrocarbon gases emitted by the li ion battery through reactions like:
$$ \text{H•} + \text{F•} \rightarrow \text{HF} $$
$$ \text{OH•} + \text{F•} \rightarrow \text{HOF} \text{ (or other products)} $$
This chain termination effect can suppress combustion, leading to lower peak fire temperatures and pressures. Experimental studies have shown that immersion cooling can confine TR to a single cell within a module, effectively blocking propagation.
The overall heat balance during a TR incident in an immersed system must account for the coolant’s heat absorption capacity, both sensible and latent:
$$ Q_{TR} = (m c_p \Delta T)_{coolant} + (m h_{fg})_{coolant} + Q_{loss} $$
where \(Q_{TR}\) is the total heat released by the failing li ion battery. A coolant with high \(c_p\) and \(h_{fg}\) can absorb more energy, mitigating the temperature rise in adjacent cells.
Conclusions and Future Perspectives
Immersion cooling represents a transformative approach to li ion battery thermal management, offering unparalleled heat dissipation rates, superior temperature uniformity, and a built-in safety barrier against catastrophic failure. Research has validated the efficacy of both single-phase (mineral/silicone oils) and two-phase (fluorinated fluids) systems under various operational stresses for the li ion battery.
The future development of immersion cooling technology for li ion battery applications can be charted along three primary research vectors:
- Advanced Coolant Engineering: Developing next-generation dielectric fluids with optimally balanced properties—high thermal conductivity, low viscosity, high specific/latent heat, robust dielectric strength, and proven long-term chemical compatibility with li ion battery materials. Nanofluid formulation and stability remain key challenges.
- System-Level Integration and Optimization: Moving beyond fundamental proof-of-concept to design compact, lightweight, and energy-efficient total immersion systems. This includes optimizing flow architectures (manifolds, baffles), integrating power electronics cooling, developing efficient condensers for two-phase systems, and conducting thorough techno-economic and lifecycle analyses for automotive and grid-scale li ion battery storage applications.
- Deepening the Understanding of Safety Mechanisms: Conducting fundamental and applied research to quantitatively elucidate the multi-physics phenomena during thermal runaway suppression. This involves detailed experimental and modeling studies of the complex interactions between decomposing li ion battery materials, ejected gases, and the coolant at extreme temperatures to validate and refine the proposed chemical inhibition mechanisms.
While significant progress has been made in laboratory settings, the path to widespread commercialization of immersion cooling for li ion battery systems requires rigorous validation of long-term reliability, cost-effectiveness, and performance under real-world driving and cycling conditions. Its application may see earlier adoption in stationary energy storage systems where size and weight constraints are less stringent than in EVs, providing a crucial pathway for maturing this potent technology to safeguard the future of high-performance and high-safety li ion battery systems.