Comparative Analysis of Oxygen Index in Different Thermal Insulation Materials

The oxygen index (limiting oxygen index, LOI) is one of the most widely used small-scale indicators for evaluating the flammability of thermal insulation materials. This article presents a comparative analysis of LOI values for common insulation products—including EPS, XPS, PUR, PIR, phenolic foam, elastomeric NBR/PVC foam, polyolefin foam, and mineral wool—based on ASTM D2863 and ISO 4589-2 test methodology. It also explains how oxygen index data support material selection, factory quality control, and fire-safety compliance in building and industrial applications.

Technical Overview

The oxygen index, also known as the limiting oxygen index (LOI), is defined as the minimum volumetric concentration of oxygen in a flowing oxygen–nitrogen mixture that will just sustain candle-like combustion of a vertically mounted specimen. Because ambient air contains approximately 20.9% oxygen by volume, materials with an LOI below 21% will generally continue to burn in normal air, while materials with higher values tend to self-extinguish once the ignition source is removed.

Oxygen index testing of insulation materials is normally performed to ASTM D2863, ISO 4589-2, or the technically equivalent Chinese standard GB/T 2406.2. Although LOI alone does not determine a fire classification under GB 8624 or EN 13501, it is a fast, low-cost, and highly reproducible screening method widely used for quality control, flame-retardant formulation development, and comparative ranking of competing materials. As a general rule of thumb:

• LOI below 21%: readily combustible, sustains burning in normal air.
• LOI of 21–27%: slow-burning or partially self-extinguishing behavior.
• LOI above 27%: generally regarded as flame-retardant; many building specifications require at least 32% for Class B1 organic insulation products.

Testing Samples

Expanded Polystyrene (EPS)

EPS is a closed-cell, bead-molded foam with a typical density of 15–30 kg/m³. Plain EPS is highly combustible, with an LOI of roughly 18–20%. Flame-retardant grades modified with polymeric brominated additives can reach an LOI of 30% or higher, which is the threshold commonly required by Chinese product standards for flame-retardant EPS boards.

Extruded Polystyrene (XPS)

XPS is a continuous extruded board with a fine closed-cell structure and a typical density of 25–45 kg/m³. Standard flame-retardant grades typically achieve an LOI of 26–30%. Like EPS, XPS melts and shrinks when exposed to flame, so LOI results must always be interpreted alongside melt-dripping behavior.

Rigid Polyurethane (PUR) Foam

PUR foam, used for board stock, spray foam, and pipe insulation, has a typical density of 30–60 kg/m³. Unmodified PUR burns readily, while phosphorus- and halogen-containing flame-retardant formulations typically reach an LOI of 24–28%.

Polyisocyanurate (PIR) Foam

PIR is produced with a higher isocyanate index, which forms thermally stable isocyanurate rings and promotes char formation. With a typical density of 35–60 kg/m³, PIR generally achieves an LOI of 28–40%, clearly outperforming conventional PUR in oxygen index testing.

Phenolic Foam (PF)

Phenolic foam is based on a highly cross-linked, aromatic resole resin with an exceptionally high char yield. At a density of 40–80 kg/m³, it typically achieves an LOI of 38–50%, the highest among common organic insulation foams, combined with low smoke emission and no flaming drip.

Elastomeric NBR/PVC Foam

Flexible rubber-plastic foam, based on nitrile-butadiene rubber blended with PVC, is widely used on HVAC piping and ductwork at densities of 40–80 kg/m³. Flame-retardant grades commonly achieve an LOI of 30–38%, and Class B1 products in the Chinese market are typically specified with a minimum LOI of 32%.

Polyolefin (PE/EPE/XLPE) Foam

Polyethylene foams, including chemically cross-linked XLPE, have densities of roughly 20–40 kg/m³. Untreated PE burns readily with an LOI of about 18%, while flame-retardant cross-linked grades can reach 26–30%.

Mineral Wool and Glass Wool

Rock wool and glass wool are inorganic fiber products that do not contain a combustible polymer matrix. They are classified as non-combustible (A1) under EN 13501 and GB 8624 based on the 750 °C furnace test, so an LOI value is not normally reported; in practice they exhibit no sustained ignition under any oxygen concentration.

Testing Items

The core test item for this comparison is the determination of the limiting oxygen index at ambient temperature (23 °C). Depending on project or certification requirements, the test program may be extended with related items:

• Limiting oxygen index (LOI) to ASTM D2863 / ISO 4589-2 / GB/T 2406.2.
• Temperature index to ISO 4589-3, measuring LOI at elevated temperatures up to 200–300 °C.
• Horizontal and vertical burning tests to UL 94 or GB/T 2408 for complementary ranking.
• Smoke density, single burning item (SBI), and cone calorimeter tests where a formal fire classification is required.

Testing Methods

Apparatus and Specimens

A standard oxygen index instrument consists of a heat-resistant glass chimney, a specimen holder, an oxygen–nitrogen gas mixing and flow control system with calibrated flowmeters or mass flow controllers, and a small pilot flame igniter. For rigid self-supporting insulation foams, specimens are typically cut to 100 mm × 10 mm × 10 mm, with molded skins removed so that the test reflects the core material.

Conditioning

Specimens are conditioned at 23 ± 2 °C and 50 ± 5% relative humidity for at least 40 hours before testing. For hygroscopic or open-cell materials, extended conditioning may be specified, because absorbed moisture can artificially inflate the measured oxygen index.

Test Procedure

The specimen is clamped vertically in the chimney and purged with an oxygen–nitrogen mixture at a total linear gas velocity of approximately 40 mm/s. The pilot flame is applied to the top of the specimen following the ignition sequence defined in the standard. The operator records whether the specimen burns at least 50 mm within 180 s or self-extinguishes before that criterion is reached. The oxygen concentration is then adjusted—coarse steps for scouting and fine steps of 0.2% near the critical point—and the sequence is repeated until the pass/fail transition is bracketed and verified with additional specimens.

Calculation and Reporting

The oxygen index is calculated as the volume fraction of oxygen in the gas mixture: LOI = 100 × V(O2) / [V(O2) + V(N2)]. The final value is reported to the nearest 0.1% together with specimen dimensions, orientation, density, and conditioning. Within a single laboratory, repeatability is typically better than ±0.5 to 1 percentage point when gas calibration and specimen preparation are tightly controlled.

Comparative Analysis of Results

Consolidated laboratory experience and published product data allow the common insulation materials to be ranked as follows by typical room-temperature oxygen index:

• Untreated PE/EPE foam: approximately 18%.
• Plain EPS: 18–20%.
• Rigid PUR foam (flame-retardant): 24–28%.
• XPS (flame-retardant): 26–30%.
• Flame-retardant EPS: 30% or higher.
• Elastomeric NBR/PVC foam: 30–38%.
• PIR foam: 28–40%.
• Phenolic foam: 38–50%.
• Mineral wool and glass wool: non-combustible, no sustained ignition.

Several conclusions emerge from this comparison. First, polymer chemistry dominates the result: the aromatic, highly cross-linked structures of phenolic resin and PIR generate protective char layers that materially raise the oxygen index, whereas linear aliphatic chains in EPS and PE offer no inherent char-forming mechanism. Second, flame-retardant additives can raise the LOI of ordinary foams substantially—for example, lifting EPS from below 20% to above 30%—but their effect can decline after thermal aging, UV exposure, or water immersion, so retention testing is advisable for long-life building products. Third, the oxygen index falls as ambient temperature rises, which is why elevated-temperature index testing to ISO 4589-3 provides a more realistic ranking for fire-exposure scenarios. Finally, a high LOI does not by itself guarantee fire safety; parameters such as heat release rate, smoke production, flaming drips, and flame spread must be assessed with cone calorimeter, SBI, or full classification tests before final material selection.

Applications

Oxygen index comparison data directly support engineering decisions across the thermal insulation industry:

• External wall and roof insulation systems, where codes restrict combustible foams and where PIR, phenolic, or mineral wool solutions are preferred in fire compartments.
• HVAC and chilled-water pipe insulation, where flexible NBR/PVC foam with an LOI of 32% or higher is routinely specified for Class B1 performance.
• Cold-chain facilities, LNG and petrochemical plants, and cleanrooms, where insulation must combine low thermal conductivity with verified flame retardance.
• Factory quality control, where every production batch of flame-retardant foam is checked against a contractual LOI minimum before release.
• Research and development of halogen-free flame-retardant formulations, where LOI serves as a fast iteration-screening tool before full-scale fire testing.
• Third-party certification and type testing under GB 8624, EN 13501, and product standards for CE marking or domestic market access.

FAQ

What oxygen index value is generally required for Class B1 organic insulation in China?

While the formal B1 classification under GB 8624 is based on full-scale burning behavior, the relevant product standards commonly require flame-retardant EPS to reach an LOI of at least 30% and Class B1 elastomeric NBR/PVC foam to reach at least 32%. Individual project specifications may impose stricter limits.

Does a high oxygen index mean a material will not burn in a real fire?

No. LOI is a small-scale, well-ventilated ignition test. In a developed fire, elevated temperatures and high radiant heat fluxes reduce the effective oxygen index, and factors such as smoke toxicity, flame spread, and dripping are not captured by LOI alone.

Why is no oxygen index reported for mineral wool or glass wool?

These are inorganic fiber materials with no combustible polymer phase. They pass the 750 °C non-combustibility test and are classified A1, so LOI testing is not applicable and manufacturers report "no ignition" instead of a numerical value.

How does the oxygen index differ from UL 94 ratings?

The oxygen index is a quantitative measure—the minimum oxygen concentration that supports combustion—whereas UL 94 is a pass/fail assessment of after-flame time, afterglow, and flaming drips in normal air. The two methods are complementary, and materials are often qualified against both.

How often should oxygen index testing be repeated?

For production control, LOI is typically verified on every batch or on a defined sampling frequency. Retesting is also required after any change to the resin system, flame-retardant package, or foaming formulation, and periodically during annual type testing to confirm retention of flame-retardant performance.