Oxygen Index Determination of Thermal Insulation Materials: Principles, Standards, and Applications

The oxygen index test, also known as the limiting oxygen index (LOI) determination, is a fundamental small-scale flammability method for quantifying how readily thermal insulation materials such as polyurethane foam, polystyrene board, phenolic foam, and elastomeric NBR/PVC foam will burn. Performed in accordance with ISO 4589-2, ASTM D2863, or GB/T 2406.2, the method measures the minimum oxygen concentration in an oxygen–nitrogen mixture that just sustains candle-like combustion, providing a reproducible index for flame-retardant grade verification, factory quality control, and building fire-safety compliance. This article explains the test principle, sample requirements, procedural steps, result interpretation, and typical applications for insulation manufacturers and fire-testing engineers.

Technical Overview

The limiting oxygen index (LOI), also referred to as the oxygen index (OI) or critical oxygen index, is defined as the minimum volumetric concentration of oxygen in a flowing oxygen–nitrogen mixture that will just support sustained flaming combustion of a vertically mounted specimen under specified laboratory conditions. The result is expressed as a percentage and calculated as LOI = 100 × V(O₂) / [V(O₂) + V(N₂)], where V(O₂) and V(N₂) are the volumetric flow rates of oxygen and nitrogen at the limiting concentration.

Because ambient air contains approximately 20.9% oxygen, the LOI gives an immediate indication of burning tendency in air. Materials with an LOI below about 21 tend to continue burning once ignited, while materials with an LOI of 27 or above are generally regarded as self-extinguishing or flame-retardant. For thermal insulation materials—many of which are low-density organic foams with high surface area and a correspondingly high fire load—a verified oxygen index value is a primary acceptance parameter in product standards, project specifications, and factory production control.

The principal standards governing oxygen index determination include:

• ISO 4589-2 – Plastics – Determination of burning behaviour by oxygen index – Part 2: Ambient-temperature test (the international reference method, also adopted as JIS K 7201-2)
• GB/T 2406.2 – Chinese national standard technically aligned with ISO 4589-2, widely applied to insulation products in the Chinese market
• ASTM D2863 – Standard Test Method for Measuring the Minimum Oxygen Concentration to Support Candle-Like Combustion of Plastics (Oxygen Index)
• ISO 4589-3 – Elevated-temperature oxygen index testing for assessing burning behaviour above ambient temperature
• GB 8624 and EN 13501-1 – Fire classification frameworks for building products, in which oxygen index data support material development and quality control, while final classification relies on scenario-based reaction-to-fire tests

Testing Samples

Oxygen index determination is applicable to self-supporting or supported specimens cut from insulation products in their delivered state. The main material families submitted for testing are summarized below.

Rigid Polyurethane (PUR) and Polyisocyanurate (PIR) Foams

Specimens are cut from insulation boards, sandwich panel cores, and spray-applied foams. Because foam cell structure is anisotropic, specimens are typically prepared both parallel and perpendicular to the foam rise direction and the orientation is reported. Unmodified PUR foams typically show LOI values in the range of 21–26, while PIR formulations commonly achieve 28–38 due to their highly charring, isocyanurate-based chemistry.

Expanded and Extruded Polystyrene (EPS and XPS)

Unmodified polystyrene foams have LOI values of approximately 17–19 and burn readily in air. Flame-retarded grades containing polymeric or halogenated flame retardants, often in combination with synergists, typically reach LOI values of 24–30. Since flame retardants in polystyrene can migrate or thermally age, oxygen index testing is frequently repeated after accelerated aging.

Phenolic Foam

Phenolic insulation foams exhibit inherently strong char formation and typically achieve LOI values of 35–45 or higher, making oxygen index testing a straightforward confirmation of their flame-retardant character.

Elastomeric Rubber–Plastic (NBR/PVC) Foam

Flexible foam sheets and preformed tubing used for HVAC, chilled water, and refrigeration piping are tested using a U-shaped frame or similar support to keep thin, compliant specimens upright. Product specifications for these materials commonly require LOI values of 32 or above.

Polyolefin and Cross-Linked Foams

Polyethylene (PE), EVA, and chemically or physically cross-linked polyolefin foams have very low natural LOI values (17–20) and are upgraded to 25–30 with halogen-free or halogenated flame retardant systems; oxygen index testing verifies both initial and retained flame retardancy.

Fibrous and Inorganic Insulation

Glass wool and rock (mineral) wool products are non-combustible and classified in the highest fire classes without combustion-based index testing. Oxygen index testing is not applicable to such inorganic fibrous materials, and their fire performance is demonstrated by non-combustibility and calorific value tests instead.

Testing Items

A complete oxygen index determination performed by accredited third-party laboratories or in-house fire laboratories typically delivers the following items:

• Limiting oxygen index (LOI) at ambient temperature, reported to one decimal place, together with specimen dimensions and orientation
• Elevated-temperature oxygen index, where required, per ISO 4589-3
• Documented observations of burning behaviour, including afterflame and afterglow, melting, flaming drips, shrinkage, charring, and smoke release
• Verification data for gas purity, flow calibration, and chimney gas velocity
• Comparison of results against specified acceptance limits, such as LOI ≥ 26, ≥ 28, or ≥ 32, as called up by the governing product standard or contract

Testing Methods

Specimen Preparation and Conditioning

Under ISO 4589-2 and GB/T 2406.2, rigid self-supporting specimens are typically 80–150 mm long, 10 ± 0.5 mm wide, and 4 ± 0.25 mm thick, or at the actual product thickness where representative. ASTM D2863 specifies specimens 70–150 mm long, 6.5 ± 0.5 mm wide, and 3 ± 0.25 mm thick. A gauge mark is placed 50 mm from the top of each specimen to define the burn-length criterion. Specimens are conditioned at 23 ± 2 °C and 50 ± 5% relative humidity for at least 88 hours unless a different conditioning state (for example, dry or as-received) is specified. A sufficient number of specimens, typically 10–20 per condition, is prepared to complete the concentration bracketing sequence.

Apparatus

The test assembly consists of a vertical glass chimney of 75–100 mm internal diameter and approximately 450–500 mm height, mounted on a base packed with glass beads to ensure homogeneous mixing of the incoming gases. High-purity oxygen and nitrogen, typically at least 99.5% purity, are metered through calibrated mass flow controllers or flowmeters and combined so that the total upward gas velocity in the chimney is 40 mm/s at the test concentration. A specimen holder clamps the specimen vertically with its top edge at least 100 mm below the top of the chimney; flexible foam specimens are supported in a frame. Ignition is provided by a small propane or butane diffusion flame applied downward onto the top edge of the specimen, and a timer is used to measure burning time. An oxygen analyzer may be used to verify the delivered mixture composition.

Test Procedure

The general procedure under ISO 4589-2, GB/T 2406.2, and ASTM D2863 is as follows:

• Mount the specimen vertically, purge the chimney with the oxygen–nitrogen mixture at the initial concentration, and allow the gas flow to stabilize for at least 30 seconds.
• Apply the ignition flame to the top of the specimen in short intervals, typically up to 5 seconds of application alternating with brief withdrawal, until the specimen ignites, without exceeding the permitted total ignition time.
• Record whether the specimen burns, defined as continued flaming for at least 3 minutes after removal of the igniter or as the flame front consuming at least 50 mm of specimen length (reaching the gauge mark); otherwise the specimen is recorded as extinguishing.
• Adjust the oxygen concentration in systematic steps—coarse steps of 0.5–1% for initial bracketing, refined to steps of approximately 0.2%—to locate the transition between burning and extinguishing behaviour, following the ascending/descending sequence defined in the standard.
• Repeat on replicate specimens to complete the required response series, noting any dripping, melting, or distortion that affects judgement.

Calculation and Expression of Results

The limiting oxygen index is calculated as the volume fraction of oxygen in the oxygen–nitrogen mixture at the limiting condition and is reported to the nearest 0.1%. The test report must state the standard used, specimen dimensions, thickness, and orientation relative to the foam rise or machine direction, conditioning details, ignition procedure, gas velocity, observed burning phenomena such as dripping or afterglow, and the oxygen concentration step size employed. Where replicate precision is required, the mean value and, where applicable, a measure of scatter are reported.

Factors Affecting Results

Several parameters influence oxygen index values and must be controlled or disclosed to ensure comparability:

• Specimen thickness and orientation, since cellular foams exhibit anisotropic burning along and across the rise direction
• Foam density, cell size distribution, and structural uniformity within the product
• Moisture content and conditioning history, particularly for open-cell and hygroscopic materials
• Gas purity, flowmeter calibration, chimney gas velocity, and apparatus temperature
• Operator judgement in applying the 3-minute and 50 mm burn criteria and in handling melting or dripping behaviour
• Aging of flame-retardant additives, including volatilization and migration losses, which can reduce the LOI of finished products over time

Applications

Oxygen index data serve multiple technical and commercial purposes across the insulation industry:

• Verification of flame-retardant grades of PUR/PIR, XPS, EPS, phenolic, and elastomeric foam products against product standards and project specifications
• Factory production control and periodic type testing on continuous insulation manufacturing lines
• Research and development screening of flame-retardant formulations, including phosphorus-based, halogenated, mineral filler, and intumescent systems
• Supporting technical documentation for fire performance dossiers under building regulations, alongside scenario tests required by classification standards such as GB 8624 and EN 13501-1
• Material qualification for transportation applications, including rail and marine equipment, where specifications frequently cite minimum LOI acceptance limits
• Supplier qualification, incoming inspection, comparative material selection, and failure or dispute analysis

FAQ

What oxygen index value qualifies insulation as flame retardant?

As a practical guideline, materials with an LOI of 27 or above are self-extinguishing in normal air, and many specifications for B1-type elastomeric, phenolic, and PIR insulation products require LOI values of 32 or higher. However, the applicable limit always depends on the governing product standard or contractual specification, which must be consulted before setting acceptance criteria.

How does oxygen index testing differ from other fire test methods?

The oxygen index test is a small-scale, well-controlled method that measures ease of sustained ignition in a defined atmosphere with excellent repeatability and low specimen consumption. It does not simulate a developed fire. It is therefore used alongside scenario-based methods such as the cone calorimeter (ISO 5660), the single burning item test used in European and Chinese classification schemes, smoke density tests, and small-flame ignition classifications such as UL 94.

Why might two laboratories report different LOI values for the same product?

Differences commonly arise from specimen thickness and orientation, conditioning state, gas flow calibration, chimney velocity, operator application of the burn criteria, and the age of flame-retardant additives in the sampled batch. Full compliance with the reporting requirements of ISO 4589-2 or ASTM D2863, including orientation and thickness disclosure, greatly improves inter-laboratory agreement.

Does specimen thickness affect the oxygen index result?

Yes. Thicker specimens generally show somewhat lower LOI values because heat losses are proportionally smaller and combustion feedback is stronger. For this reason, standards fix or require reporting of the test thickness, and results for different thicknesses should not be directly compared.

Can an LOI result alone be used to classify a building product?

No. Fire classification under GB 8624 and EN 13501-1 is based on scenario tests covering heat release, flame spread, smoke production, and flaming droplets. Oxygen index results are used for development, production control, and specification compliance rather than as a standalone classification tool.

Can the oxygen index be measured at elevated temperatures?

Yes. ISO 4589-3 defines an elevated-temperature procedure in which the chimney atmosphere is heated. LOI values generally decrease as temperature rises, so this variant is valuable when insulation will be exposed to sustained service temperatures above ambient.