Home › Blog › EV & Battery › Thermal Runaway & Fire Protection › Barrier layering
Thermal Isolation vs. Fire Protection: How to Layer an EV Thermal-Runaway Barrier Stack
The most common specification error in thermal-runaway protection is asking one material to be two materials. Slowing heat conduction from a failing cell to its neighbor and surviving flame, hot gas, and ejected particles on a vent path are different physics, rewarded by different properties, and led by different material families. The parent application page splits every barrier request into insulate, flame-face, or both, and answers "both" with a laminated stack rather than a compromise sheet.
This article explains the split, what UL 2596's torch-and-grit method actually characterizes, why compression silently changes a barrier's thermal performance, and how a layered barrier is built and converted as one part.
Request a quote for your project →
Two jobs, two physics
Thermal isolation is a conduction problem. The barrier's job is to slow the flow of heat from one cell to the next so that a single-cell event has less chance of driving its neighbor past its own runaway threshold. The governing property is low through-thickness thermal conductivity, evaluated at the thickness and compression the layer actually has in service. The materials that lead are the low-conductivity insulators: aerogel blanket above all in thin gaps.
Fire protection is a survival problem. On a vent path, a module wall, or a pack lid, the barrier faces flame, hot gas, and entrained particles moving fast, and its job is to not erode, not burn through, and keep shielding what is behind it. The governing properties are flame and erosion resistance at very high temperature, where the inorganic and purpose-built families lead: mica barrier sheet, glass-fiber paper, and the engineered EV firewall constructions.
What UL 2596 actually tests, and why vent shields care
UL 2596 is a test method for the thermal and mechanical performance of battery enclosure materials, often described as a torch-and-grit test: it exposes a material to flame together with entrained particles, approximating the combined heat and ejecta a barrier sees when a cell vents. That combination is the point. A vent-path or lid shield must resist erosion and burn-through from hot, fast-moving gas and particles, not just elevated temperature, and materials that look equivalent under a static flame can separate sharply once the grit is added.
A vent shield is therefore chosen for its behavior under a method like UL 2596, confirmed on the manufacturer's data sheet for the specific construction; the demonstrated result is a property of the tested material and construction, and no rating value is claimed here for any grade.
Above the material level sit the assembly methods: UL 9540A evaluates how a thermal-runaway event propagates through an energy-storage system, and GB 38031 includes a thermal-propagation requirement for EV traction batteries. Both apply to the tested cell, module, pack, or system. A barrier helps an assembly meet them; it does not pass them on its own.
The compression variable that bench data hides
A cell-to-cell barrier operates at whatever thickness the stack leaves it, and that thickness moves: cells breathe reversibly each cycle and swell irreversibly over life. A compressible barrier compressed to a different thickness has different effective thermal performance, so the honest question is never the free-thickness bench value but the behavior at the in-service compressed thickness across the life of the stack.
Practically, that means the gap range and the compression the barrier sees belong on the drawing, and a stack that must both cushion and insulate often pairs a compliant layer (high-temperature silicone foam) with an insulating or flame-facing layer rather than asking either to do the whole job.
A Layered Cell-to-Cell Barrier: Each Ply Does One Job
Qualitative illustration of the layering principle from the H-O thermal-runaway application page; ply order, count, and materials are design-specific and confirmed per the manufacturer's data sheets and your assembly testing. No thickness or performance values are represented.
Layering by location: how the five jobs map to stacks
- Cell-to-cell isolation. The thinnest, most constrained slot. Aerogel blanket leads on conduction; where the layer must also cushion the breathing stack, a silicone-foam ply joins it; a facing film or edge encapsulation controls friability.
- Module and pack propagation barriers. Purpose-built EV firewall barrier sheet and the related thermal and fire barrier constructions, sized to the wall or interface geometry.
- Vent-path and lid fire shields. The UL 2596-relevant duty: erosion and burn-through resistance under flame plus particles. Mica barrier sheet and engineered firewall constructions lead.
- Busbar and high-temperature electrical barriers. Where flame duty meets an electrical role, mica and glass-fiber paper serve both, with the dielectric requirement confirmed separately per grade.
- Compartment and enclosure liners. Larger-area inorganic papers and sheets cut to the compartment geometry, often as the inner skin of a laminated panel.
Converting a layered barrier as one part
A layered barrier only works on the line if it arrives as one part. H-O laminates the plies (compliant foam, insulating core, inorganic flame face, facing films), encapsulates cut edges where a friable layer needs containment, die-cuts or laser-cuts the finished construction to the footprint with its cut-outs, and kits the geometries a pack uses. Mica's rigidity points it at laser and waterjet cutting; aerogel's dust points it at faced and encapsulated constructions.
The assembly drawing should call the stack as a single part number with the ply order stated, which is also what makes the engineering review meaningful: each ply can be checked against its own data sheet for the job it carries.
Failure modes a layered specification prevents
| Failure | Root cause | The layering fix |
|---|---|---|
| Neighbor cell heats despite a "fire" sheet | Flame-face material asked to do the conduction job | Add an insulating core; assign conduction to the insulator |
| Vent shield erodes | Insulator chosen on conductivity, facing ejecta | Flame-facing inorganic ply on the exposure side, chosen against UL 2596-type behavior |
| Barrier underperforms in service | Free-thickness bench value used; in-service compression ignored | State the gap and compression range; confirm thickness-dependent behavior on the TDS |
| Dust on the assembly line | Friable layer cut without facing or encapsulation | Faced construction or edge encapsulation, specified at the converting stage |
| Electrical surprise at a busbar | Dielectric role assumed from the thermal role | Call out the electrical requirement; verify the grade's dielectric data separately |
How H-O Products helps
H-O Products is a precision die-cut converter. We turn the insulate-versus-flame-face split into a converted construction: ply selection direction across aerogel, firewall sheet, mica, glass-fiber paper, and silicone foam; lamination, facing, and edge encapsulation; laser, waterjet, die-, and kiss-cutting to the footprint; and kitting for the line, under our ISO 9001:2015 certified quality management system with material traceability and lot-level data-sheet records.
Materials are evaluated against and support compliance with the cited methods through the manufacturer's data sheet; the propagation outcome under UL 9540A or GB 38031 belongs to your tested assembly, and no pass or number is claimed for any raw material.
Frequently asked questions
Why not just use a thicker single material instead of a laminate?
Because the two jobs reward different properties and thickness is usually rationed. A thicker insulator still erodes if it faces ejecta; a thicker flame sheet still conducts steadily into the next cell; and the cell-to-cell gap rarely has thickness to spare. A laminate assigns each job to the ply that wins it, which usually beats any single material at the same total thickness. The stack is converted and supplied as one part, so the line never sees the complexity.
Is UL 2596 a pass/fail certification for the barrier?
No. UL 2596 is a test method that characterizes how a material and construction behave under flame with entrained particles, the combined duty a vent shield sees. The output is measured behavior for the tested construction, reported by the material manufacturer; it is not a certification a converted part carries, and H-O states no rating value. Use it to compare candidate constructions for an exposure-side ply, then confirm at assembly level.
How does cell breathing interact with the barrier stack?
The stack thickness changes as cells breathe and swell, and a compressible barrier's thermal performance is thickness- and compression-dependent. A stack that must follow the cells pairs a compliant ply (high-temperature silicone foam) with the insulating or flame-facing plies, so the recovery job and the barrier jobs are separated. The drawing should state the gap range and the compression the barrier sees at both ends of life.
Who owns the propagation result, the material or the pack?
The pack. UL 9540A and GB 38031 evaluate propagation in the tested system or vehicle battery, with every design detail participating: cell chemistry, spacing, venting, cooling, and the barriers. A well-chosen laminated barrier raises the assembly's chances, but the demonstrated result belongs to your tested cell, module, pack, or system under your specific design.
Have a barrier stack-up reviewed
Send the location, the job split (insulate, flame-face, or both), the gap and compression range, and any electrical role, and engineering will propose a ply construction and converting approach.
Samples typically ship in 3 to 5 business days for common die-cut configurations; standard production runs about two weeks; special orders are quoted to the job.
Family-owned in Winsted, Connecticut since 1971 · ISO 9001:2015 certified · OEMs worldwide send us a drawing — and get back their part.
Related reading
IndustryEV & Battery hubRead →
Request a quote for your EV thermal-runaway & fire protection part
Send H-O your drawing or describe the part—engineering reviews every RFQ and converts samples on your geometry.
Last updated · Reviewed by H-O Products engineering (ISO 9001:2015 certified converter, Winsted, CT).
References
- UL 2596 (thermal and mechanical performance of battery enclosure materials: flame plus entrained particles).
- UL 9540A (thermal-runaway propagation, energy storage systems) and GB 38031 (EV traction battery safety, incl. thermal propagation), assembly-level methods.
- ASTM E162 / ASTM E662 (flammability and smoke characterization); UL 94 (classifications); UN 38.3 (transport, pack level).
- Manufacturer technical data sheets: aerogel blanket, EV firewall barrier sheet, mica barrier sheet, glass-fiber paper, high-temperature silicone foam.
This article explains the insulate-versus-flame-face design split and barrier-layering reasoning from the H-O thermal-runaway application page. The figure is qualitative; ply order and materials are design-specific, no performance value is stated for any grade, and grade behavior is confirmed on the manufacturer's current technical data sheet at your thickness and compression. Propagation outcomes belong to the tested assembly.
All parts are custom manufactured to order; minimum production run quantities vary by material and part. H-O is a die-cut converter, not a raw-material producer.