Foam Plastics Oxygen Index (LOI) Test: Principles, Standards, and Testing Procedures

The oxygen index test, also known as the limiting oxygen index (LOI) test, is the most widely used method for evaluating the flame-retardant performance of foam plastics such as EPS, XPS, and polyurethane foam. This article explains the test principle, apparatus, specimen preparation, and step-by-step procedure in accordance with ISO 4589-2, ASTM D2863, and GB/T 2406.2, helping manufacturers and engineers correctly interpret LOI results for quality control and product development.

Technical Overview

The oxygen index, formally defined as the limiting oxygen index (LOI), is the minimum concentration of oxygen, expressed as a percentage by volume, in a flowing mixture of oxygen and nitrogen that will just support flaming combustion of a vertically mounted specimen under specified test conditions. Because ambient air contains approximately 20.9% oxygen, a foam plastic with an LOI above 21% will not continue to burn in a normal atmosphere once the ignition source is removed, making the LOI a practical and reproducible indicator of flame retardancy.

Foam plastics deserve particular attention in flammability testing because their cellular structure gives them an extremely high surface-to-mass ratio, low thermal inertia, and abundant trapped air, all of which accelerate ignition and flame spread. The oxygen index experiment therefore serves as a primary quality-control and research tool for expanded polystyrene (EPS), extruded polystyrene (XPS), flexible and rigid polyurethane foams, polyethylene foam, PVC foam, and phenolic foam.

The test is governed by an internationally harmonized family of standards:

• ISO 4589-2 (Plastics — Determination of burning behaviour by oxygen index — Part 2: Ambient-temperature test), the international reference method.
• GB/T 2406.2, the Chinese national standard that is technically identical to ISO 4589-2 and is the most commonly cited method in Chinese product specifications.
• ASTM D2863, the North American method for measuring the minimum oxygen concentration to support candle-like combustion of plastics.
• ISO 4589-3 and GB/T 2406.3, which extend the determination to elevated-temperature conditions, reflecting the fact that LOI decreases as temperature rises.

Accredited third-party laboratories routinely perform oxygen index testing under these standards for building-material certification, transportation compliance, and research purposes.

Testing Samples

Rigid Foam Plastics (EPS, XPS, Phenolic Foam)

For rigid cellular plastics, specimens are typically cut to 80–150 mm in length with a width of 10 ± 0.5 mm and a thickness of 10 ± 0.5 mm, as specified in GB/T 2406.2 and ISO 4589-2. Specimens must be cut cleanly, without crushed cell structure, melted edges, or skin damage, because deformed cells alter the burning behavior. Density should be reported alongside the LOI, since foams of the same polymer chemistry but different densities can show measurably different oxygen index values. Chinese product standards such as GB/T 10801.1 for EPS thermal-insulation boards require an oxygen index of at least 30% for compliance.

Flexible Polyurethane Foam

Flexible PU foams are tested with the same nominal cross-section, usually 10 mm by 10 mm. These materials may shrink, melt, or drip during the test, and such behavior must be recorded in the report. Because flexible foams are highly anisotropic, specimens should be cut with the length axis parallel to the rise direction unless the product specification requires otherwise, and the direction of testing must be stated.

Other Foam Materials

Polyethylene (EPE), polypropylene, PVC, EVA, and silicone foams follow the same dimensional rules. Sheet materials thicker than approximately 10.5 mm must be machined down to the standard thickness, and the machining method should be noted, as cut surfaces can influence ignition. For every material category, a minimum of 15 specimens is recommended so that the oxygen concentration can be bracketed reliably during the sequential procedure.

Specimen Conditioning

All specimens must be conditioned to equilibrium at 23 ± 2 °C and 50 ± 5% relative humidity prior to testing, generally for a minimum of 88 hours in accordance with GB/T 2918 / ISO 291, unless the relevant product standard specifies otherwise. Moisture content in hydrophilic foams can measurably raise the apparent LOI, so consistent conditioning is essential for comparable results.

Testing Items

The core deliverable of the experiment is the limiting oxygen index value, expressed as a percentage by volume of oxygen. In addition, the test record should capture the following items:

• The LOI value and the oxygen concentration increment used in the final determination.
• The ignition method applied (top-surface ignition or propagating ignition).
• Observed burning behavior, including afterflame and afterglow, melting, shrinking, dripping of flaming particles, and intumescence.
• Specimen dimensions, density, orientation, and conditioning history.
• A flame-retardancy classification based on the measured LOI.

In common industrial practice, the LOI value is interpreted against the following reference thresholds:

• LOI below 21%: the material burns readily in ambient air and is considered highly flammable.
• LOI between 21% and 27%: the material shows self-extinguishing tendencies and is classified as slowly burning.
• LOI above 27%: the material is regarded as flame-retardant or difficult to ignite, which is the target range for most regulated foam applications.

Many building and transportation specifications set explicit numerical limits, such as LOI ≥ 26% or ≥ 30%, so the applicable product standard must always be consulted when judging conformity.

Testing Methods

Apparatus

The oxygen index apparatus defined in GB/T 2406.2 and ISO 4589-2 consists of the following principal components:

• A vertical heat-resistant glass chimney, typically about 500 mm high with an internal diameter of 75 mm, through which the oxygen–nitrogen mixture flows upward.
• A specimen holder that clamps the lower portion of the specimen and supports it vertically in the center of the chimney.
• Supplies of high-purity oxygen and nitrogen (minimum purity 98%) with calibrated flow-measurement devices, such as rotameters or mass flow controllers.
• An ignition source, normally a small propane or butane burner, used to ignite the top of the specimen.
• An optional oxygen analyzer for verifying the concentration of the mixed gas stream.

Test Procedure

The standard procedure proceeds as follows:

• Condition and measure the specimens, then mark a reference line 50 mm below the top edge.
• Mount the specimen vertically in the holder and position it centrally inside the chimney.
• Set the oxygen and nitrogen flows so that the total flow produces a linear velocity of 40 ± 10 mm/s through the chimney, and allow the system to purge and stabilize for approximately 30 seconds before ignition.
• Apply the ignition flame to the specimen: in Method A (top-surface ignition), the flame is applied briefly to the top surface until it ignites; in Method B (propagating ignition), the flame is applied so that the top surface ignites and the flame front begins to travel downward.
• Observe whether the specimen burns for 180 seconds or more, or whether the flame front reaches or passes the 50 mm mark. If either criterion is met, the oxygen concentration is too high; if the specimen extinguishes before either criterion, the concentration is too low.
• Adjust the oxygen concentration in progressively smaller increments, typically down to 0.2% by volume, testing fresh specimens until the boundary between burning and extinguishing is reliably bracketed.
• Calculate the LOI from the calibrated gas flows: LOI (%) = V(O₂) / [V(O₂) + V(N₂)] × 100.

Precision and Reporting

Under well-controlled conditions, repeatability within a single laboratory is typically better than ±0.5% oxygen, while reproducibility between laboratories is generally within about ±1.5% oxygen. The test report must state the standard used, specimen details, ignition method, gas flow conditions, the measured LOI, and any atypical burning behavior, since these details are essential for inter-laboratory comparison and regulatory submission.

Key factors that influence the result include specimen density and thickness, aging and conditioning, the accuracy of flow calibration, operator ignition technique, and the ambient temperature of the laboratory. For this reason, comparative evaluations should always be performed on specimens of matched density and geometry, ideally at the same accredited third-party laboratory.

Applications

The oxygen index experiment is applied throughout the foam plastics value chain:

• Building insulation: verifying the flame-retarded grade of EPS, XPS, and polyurethane insulation boards required for fire-performance classification of building products under systems such as GB 8624.
• Furniture and bedding: screening flexible polyurethane foams and rebonded foams for compliance with flammability requirements in mattresses, upholstered seating, and public-venue furnishings.
• Transportation: qualifying foam components in railway vehicle interiors, automotive cabins, ship fittings, and aircraft applications, where LOI values of 28–35% are frequently specified by procurement specifications.
• Electrical and appliance insulation: evaluating rigid PU foam used in refrigerated appliances, panel cores, and cable pathway sealing foams.
• Research and development: ranking the effectiveness of flame-retardant formulations, including phosphorus-based, nitrogen-based, halogen-based, and mineral-filled systems, and quantifying synergistic combinations in polymer foam matrices.
• Incoming quality control: manufacturers use rapid LOI screening to verify that supplied foam batches meet contractual flame-retardancy limits before production release.

FAQ

What oxygen index value is considered flame retardant for foam plastics?

As a general rule, foams with an LOI above 27% are considered flame-retardant, while values above 30% satisfy the most demanding building-insulation specifications. However, specific product standards, such as those for EPS or XPS insulation boards, must be checked because they define their own minimum limits.

How does the oxygen index test differ from the UL 94 test?

The two methods answer different questions. The oxygen index test measures the minimum oxygen concentration needed to sustain candle-like combustion in a controlled atmosphere, giving a continuous numerical value. The UL 94 test rates burning behavior, dripping, and afterflame of specimens ignited in normal air and assigns pass/fail grades such as V-0 or HF-1. The tests are complementary, and many specifications require both.

How many specimens are needed for one determination?

A minimum of 15 specimens is generally recommended by GB/T 2406.2 and ISO 4589-2, because the method requires sequential testing at several oxygen concentrations until the burning/extinguishing boundary is bracketed. Additional specimens may be needed for materials with erratic burning behavior.

Why can two laboratories report different LOI values for the same foam?

Differences usually arise from specimen density and thickness variation, conditioning history, flow-meter calibration, ignition technique, and laboratory temperature. Inter-laboratory agreement is typically within ±1.5% oxygen, which is why conformity judgments should include measurement uncertainty.

Can the oxygen index be measured at elevated temperature?

Yes. ISO 4589-3 and GB/T 2406.3 define the elevated-temperature oxygen index procedure. Because LOI decreases as temperature rises, elevated-temperature data are valuable for applications in which foams are exposed to heat in service.

Does a high LOI guarantee fire safety?

No. The oxygen index is a small-flame laboratory test that reflects ease of ignition in a controlled atmosphere, not full-scale fire behavior such as flame spread, heat release, or smoke production. A complete fire-performance assessment should combine LOI data with additional standardized tests such as cone calorimetry, flame-spread, and smoke-density methods.