In the context of accelerating global energy transition, the safety of large-scale battery energy storage systems has become a paramount concern. Among the various failure modes, the arc fault induced by thermal runaway represents one of the most critical hazards, characterized by ultra-high temperatures and energy densities that can dramatically escalate fire and explosion risks. This article provides a comprehensive overview of the research progress on arc formation mechanisms in battery energy storage systems, the influence of arc faults on thermal runaway propagation, and the development of multi-physics models for predicting arc behavior. The core objective is to elucidate the multipath coupling mechanisms driving arc formation and to propose safety design principles for enhancing the electrical integrity of battery energy storage systems.
Multipath Mechanisms of Arc Initiation in Battery Energy Storage Systems
Arc faults in battery energy storage systems are rarely the result of a single isolated event. Instead, they typically emerge from a complex interplay of thermal, electrical, mechanical, and chemical factors. The fundamental mechanisms can be categorized into three primary pathways:
| Mechanism Pathway | Primary Cause | Physical Process |
|---|---|---|
| Insulation Failure-Induced | Mechanical deformation, vibration, or assembly stress | Exposure of bare metal conductors reduces safe clearance below the critical breakdown distance (approx. 1 mm/kV in air), leading to gas dielectric breakdown. |
| Thermal Runaway Product-Triggered | Ejection of high-temperature gases, particles, and electrolyte vapor | Conductive particles and vaporized electrolyte alter the local dielectric environment, reducing insulation strength by 20% – 50%, inducing non-uniform field distortion. |
| Structural Degradation-Accumulated | Loose electrical connections and chemical corrosion | Progressive insulation breakdown over the system lifecycle leads to partial discharge in high-voltage circuits (e.g., DC busbars, contacts), evolving into a sustained arc. |
The coupling of these pathways can create a cascading disaster cycle: thermal runaway → insulation failure/electrical field alteration/conductive channel formation → arc generation → accelerated thermal propagation/combustion. This cascade is the core threat to the safety of large-scale battery energy storage systems.
Insulation Material Failure and Arc Ignition
In the safety risk chain of battery energy storage systems, the degradation of insulation materials under thermal stress is a primary precursor to arc faults. As battery energy densities and voltage platforms increase, insulation materials face more severe coupled electrical-thermal-mechanical challenges. Common materials used in battery systems–including mica paper, insulation bubble film, structural adhesives, and battery blue film–have been subjected to high-temperature tests (ramped at 10 °C/min to 400 °C) to measure resistance changes. At 200 °C, structural adhesives begin to pyrolyze; at 300 °C, they soften and lose their shape; and at 400 °C, they completely melt. The measured resistance values after treatment are shown below:
| Material | Resistance at 200 °C (Ω) | Resistance at 300 °C (Ω) | Resistance at 400 °C (Ω) |
|---|---|---|---|
| Mica Paper | 3.04 × 10⁹ | 2.91 × 10¹¹ | 5.20 × 10¹¹ |
| Insulation Bubble Film | 2.94 × 10⁹ | 1.21 × 10¹¹ | 1.59 × 10¹¹ |
| Structural Adhesive | 1.40 × 10¹² | — (softened) | — (melted) |
| Battery Blue Film | 3.17 × 10⁹ | 1.73 × 10¹⁰ | 1.40 × 10¹¹ |
At 400 °C, while the resistance values of these materials increase, their physical contraction or loss of integrity exposes charged components, directly increasing the risk of short circuits and arc ignition. This is especially critical in battery energy storage systems where high voltages (e.g., 400 V, 800 V, 1500 V) are standard. The reduction in electrical safety clearance due to insulation breakdown can cause the electric field to exceed the dielectric strength of the interposed medium, leading to gas breakdown and arc formation according to Paschen’s law:
$$V_b = \frac{Bpd}{\ln(Apd) – \ln[\ln(1 + \frac{1}{\gamma})]}$$
where \(V_b\) is the breakdown voltage, \(p\) is the gas pressure, \(d\) is the electrode gap, \(A\) and \(B\) are gas-dependent constants, and \(\gamma\) is the secondary emission coefficient. In a compromised insulation scenario, the effective gap \(d\) decreases, or the dielectric medium is altered by thermal decomposition products, drastically reducing \(V_b\).
Particulate-Induced Arc in Battery Energy Storage Systems
During thermal runaway, battery cells eject a large volume of high-temperature, flammable smoke and solid particulate matter (PM). These particles, as they flow, migrate and deposit onto high-voltage components within the system, creating potential arc paths. A detailed characterization of these particles–including mass, size distribution, elemental composition, and morphology–is essential for understanding arc initiation. The critical findings from recent studies on Li(Ni₀.₈Co₀.₁Mn₀.₁)O₂ (NCM811) batteries are summarized below.
| Parameter | Value Range | Impact on Arc Threshold |
|---|---|---|
| Particle Size (diameter) | nano-scale to >2 mm | Larger particles (>100 µm) and metallic fragments (>2 mm) drastically lower the breakdown voltage. |
| Conductivity of Fragments | Cu: 1.11 – 3.99 mΩ | Metallic fragments create low-resistance bridges, facilitating sustained arcing. |
| Mass Fraction of Fragments | ~5.78% of total ejecta | Despite low mass, they increase arc probability by over 300% compared to smaller particles. |
Experiments designed to simulate particle deposition between charged electrodes have revealed that the critical breakdown voltage for arc initiation is a function of both particle size (\(D_0\)) and electrode gap (\(d_0\)). The critical breakdown voltage (\(V_{crit}\)) for a particle-laden gap can be expressed empirically as:
$$V_{crit} \approx k \cdot f(D_0) \cdot d_0^2$$
where \(k\) is a material-dependent constant and \(f(D_0)\) is a function that decreases with increasing particle size. For de-sensitized NCM811 battery ejecta, the minimum \(V_{crit}\) was found to be approximately 99 ± 5 V for a 1 mm gap, compared to several kilovolts for a pure air gap. This reduction is primarily attributed to the graphite particles (approx. 70% of the PM by mass), which reduce the effective breakdown strength to between 0.9% and 2.3% of that of clean air. The relationship between gap distance and voltage threshold follows a quadratic positive correlation for gaps < 8 mm, whereas for gaps > 8 mm, particulate matter cannot induce arcs below 400 V. This leads to a safety design boundary for battery energy storage systems:
$$U_c \geq 4.08 L^2$$
where \(U_c\) is the voltage across the arc zone and \(L\) is the distance between electrodes. This equation provides a theoretical basis for defining minimum safe electrical clearances in battery packs to prevent particle-induced arcing, a critical design rule for high-voltage battery energy storage systems.
Electrolyte-Induced Arc Mechanisms
In battery energy storage systems for grid applications, lithium iron phosphate (LFP) batteries are widely used. During thermal runaway, these batteries eject a substantial volume of electrolyte vapor and liquid mixed with particles. The electrolyte, while having a relatively high conductivity of approximately 13.45 mS/cm (compared to 3.5 × 10⁸ mS/cm for aluminum), does not directly form a short circuit under low voltage but becomes a powerful arc initiator under high voltage (e.g., 235 V DC). Experimental observations of electrolyte-induced arcing reveal a distinct sequence:
| Stage | Time (s) Relative to Voltage Onset | Observed Phenomenon |
|---|---|---|
| 1. Initial Electrolysis | 0 – 1 | Brief flash and gas generation at electrolyte-electrode interface. |
| 2. Evaporation | 6 – 16 | Rapid evaporation of electrolyte due to Joule heating; white smoke emission. |
| 3. Pre-Arc Glow | 16 – 18 | Formation of a carbonized, black char; surface flashover begins. |
| 4. Sustained Arc | 18 | Intense arc with blinding light; temperatures reach approximately 6000 °C, causing severe electrode (Al) melting and erosion. |
The chemical degradation of the electrolyte under high voltage and temperature produces a conductive carbonaceous residue. This residue bridges the electrode gap, enabling a sustained, high-energy arc. The extreme temperature of this arc can ignite the surrounding flammable gases (H₂, CO, CH₄) produced during thermal runaway, leading to a deflagration or explosion. This mechanism is particularly dangerous in large-format LFP cells used in battery energy storage systems, where a single cell’s thermal runaway can release liters of electrolyte, coating nearby busbars and terminals and creating a perfect pathway for arc propagation.
The power dissipated in the arc can be approximated by:
$$P_{arc} = I_{arc} \cdot V_{arc}$$
where \(I_{arc}\) is the arc current and \(V_{arc}\) is the arc voltage. For a high-power system, this energy input is sufficient to vaporize metals and trigger a secondary, more severe thermal event.
Arc-Battery Interaction and Cascading Failure
Once an arc is established in a battery energy storage system, it can directly interact with adjacent cells, leading to cascading thermal runaway. The arc’s impact on a cell is highly dependent on the state of charge (SOC), system voltage, current, and electrode gap. The electrical and thermal signature of a series arc in a prismatic LFP cell has been systematically studied.
| Parameter | Effect on Arc Duration | Effect on Arc Severity |
|---|---|---|
| State of Charge (SOC) | Higher SOC → shorter time to cell failure | Higher SOC → more violent thermal runaway, larger mass loss, more flame propagation |
| System Voltage (Udc) | Higher voltage → longer arc duration (for constant current) | Higher voltage → more energy input, more material consumption |
| Loop Current (Idc) | Higher current → shorter arc duration | Higher current → more intense arcing, higher temperature, faster electrode erosion |
| Electrode Gap (L) | Larger gap → shorter arc duration (requires more power to sustain) | Larger gap → longer arc column, more energy required for maintenance |
The experimental data shows that for a 200 V DC system with 20 A current and a 0.5 mm gap, the arc duration is 20.3 seconds. Increasing the gap to 1 mm reduces the duration to 13.2 seconds. This is because the arc voltage \(V_{arc}\) increases with gap length, demanding higher power input from the circuit. The energy deposited into the battery before failure is the key metric:
$$E_{in} = \int_{0}^{t_{arc}} V_{arc}(t) \cdot I_{arc}(t) dt$$
This energy is transferred as heat to the battery’s terminals and internal structure. The heat flux from the arc can melt the battery’s safety vent, insulation seals, and current collectors (aluminum and copper), creating internal short circuits that precipitate thermal runaway. In experiments, the arc directly ignited the flammable vent gas, demonstrating that the arc is not only a consequence of thermal runaway but also a potent trigger for it.
Multiphysics Modeling of Arc in Battery Energy Storage Systems
Developing a predictive model for arc initiation and propagation within a battery energy storage system is essential for design optimization and safety assessment. The core of modern arc simulation is the solution of the magnetohydrodynamics (MHD) equations, which couple electromagnetic fields with fluid dynamics and heat transfer under the assumption of local thermal equilibrium (LTE). The governing equations include the conservation of mass, momentum, and energy:
Conservation of mass (continuity equation):
$$\frac{\partial \rho}{\partial t} + \nabla \cdot (\rho \vec{u}) = 0$$
Conservation of momentum (Navier-Stokes with Lorentz force):
$$\rho \left( \frac{\partial \vec{u}}{\partial t} + (\vec{u} \cdot \nabla)\vec{u} \right)
= -\nabla p + \mu \nabla^2 \vec{u} + \vec{J} \times \vec{B}$$
Conservation of energy (with Joule heating):
$$\rho C_p \left( \frac{\partial T}{\partial t} + \vec{u} \cdot \nabla T \right)
= \nabla \cdot (k \nabla T) + \sigma |\vec{E}|^2 – S_{rad}$$
where:
| Symbol | Description |
|---|---|
| \(\rho\) | Gas density |
| \(\vec{u}\) | Velocity vector |
| \(p\) | Pressure |
| \(\mu\) | Dynamic viscosity |
| \(\vec{J}\) | Current density vector |
| \(\vec{B}\) | Magnetic flux density vector |
| \(C_p\) | Specific heat at constant pressure |
| \(T\) | Temperature |
| \(k\) | Thermal conductivity |
| \(\sigma\) | Electrical conductivity |
| \(\vec{E}\) | Electric field vector |
| \(S_{rad}\) | Radiation heat loss term |
The electromagnetic field is described by Maxwell’s equations and Ohm’s law in the MHD approximation:
$$\vec{J} = \sigma (\vec{E} + \vec{u} \times \vec{B})$$
$$\nabla \times \vec{E} = -\frac{\partial \vec{B}}{\partial t}$$
$$\nabla \times \vec{B} = \mu_0 \vec{J}$$
Finite element models (FEM) utilizing these MHD equations have been applied to simulate arc behavior on the terminal poles of prismatic lithium-ion batteries. These simulations successfully reproduce the temperature field (peaking at over 6000 K), the magnetic flux density distribution, and the flow field of the plasma jet. Such models are crucial for understanding the impact of a sustained arc on the structural integrity of a battery module within a battery energy storage system. The model can predict the extent of electrode melting and the zone of thermal influence, allowing for the design of physical barriers or advanced fusing strategies.
However, current MHD models are limited by their assumptions. They typically assume a fully developed, stable arc and do not accurately simulate the dynamic, transient initiation phase that involves complex physics like streamer formation and dielectric breakdown. Furthermore, they often treat the plasma as a single fluid in local thermal equilibrium, which may not hold during the early stages of arc ignition or in the presence of a complex mixture of air, electrolyte vapor, and solid particles. The chemical interactions between the arc plasma and the battery’s electrolyte or structural materials (e.g., formation of copper oxide or aluminum nitride) are also generally neglected.
Conclusion and Future Directions
Arc faults in battery energy storage systems represent a critical, highly energetic failure mode that can rapidly escalate thermal runaway events into major fires or explosions. The initiation mechanisms are diverse and interconnected, involving insulation degradation at high temperatures, the deposition of conductive thermal runaway ejecta (particles and electrolyte), and the progressive structural degradation of electrical connections. The interaction between a sustained arc and a battery cell is a two-way process: the arc can trigger thermal runaway in a healthy cell, while a thermal runaway event creates the conditions for an arc to occur.
Current research has made significant strides in characterizing these mechanisms. Experimental studies have quantified the critical parameters for particle-induced and electrolyte-induced arcing, providing a scientific foundation for safety boundaries. For example, the relationship \(U_c \geq 4.08 L^2\) offers a direct design guideline for high-voltage battery energy storage systems to prevent metallic fragment-initiated arcs. Furthermore, MHD-based multiphysics models have begun to provide valuable insights into the steady-state behavior of arcs on battery terminals, mapping temperature, flow, and electromagnetic fields.
Despite this progress, critical knowledge gaps remain. The dynamic transition from a pre-breakdown event to a fully developed arc in the environment of a thermal runaway event is poorly understood. Future research should focus on the following areas:
1. Integrated “Thermal-Electrical-Mechanical-Chemical” Model Development: There is an urgent need for an intelligent simulation tool that can couple the transient thermal runaway chemistry of a cell with the electrical arc physics. This model must account for the variable dielectric properties of the gaseous medium as it changes from air to a mix of decomposition products.
2. High-Fidelity Dynamic Arc Initiation Modeling: Moving beyond steady-state MHD simulations to model the nanosecond-to-microsecond scale processes of streamer formation and dielectric breakdown in a multi-component, particulate-laden environment.
3. Real-time Arc Detection and Mitigation in Battery Energy Storage Systems: Developing advanced sensing algorithms (e.g., based on high-frequency current ripple analysis) and fast-acting protection devices (e.g., solid-state circuit breakers) specifically designed to interrupt arcs before they can cause cascading failure.
4. Arc-Triggered Gas Ignition Studies: Quantitative investigation of the conditions under which the arc plasma can ignite the flammable gas mixture (H₂, CO, CH₄) ejected from a failing cell, as this represents the core mechanism for an explosion in an enclosed battery energy storage system container.
By addressing these challenges, the research community can provide the theoretical and practical tools necessary to mitigate arc hazards, enabling the safe and reliable deployment of the next generation of high-energy-density battery energy storage systems.