Limiting Oxygen Index (LOI) Testing of Plastics: Standard Test Procedure and Technical Specification

The Limiting Oxygen Index (LOI) test is one of the most widely used bench-scale methods for quantifying the relative flammability of plastic materials. This technical article presents the complete LOI testing procedure in accordance with ISO 4589-2, ASTM D2863, GB/T 2406.2, and JIS K 7201-2, covering sample preparation, apparatus requirements, test steps, calculation, and result interpretation. It serves as a practical reference for material engineers, quality-control personnel, and accredited third-party laboratories engaged in flame-retardant material development and compliance verification.

Technical Overview

The Limiting Oxygen Index (LOI), also known as the oxygen index (OI) or critical oxygen index, is defined as the minimum concentration of oxygen, expressed as volume percent, in a flowing mixture of oxygen and nitrogen that will just support candle-like combustion of a vertically mounted plastic specimen. Because ambient air contains approximately 20.9% oxygen, a material with an LOI below 21% will generally continue to burn in air, while a material with an LOI well above 21% will tend to self-extinguish once the ignition source is removed.

The LOI method is governed by several internationally recognized standards, including ISO 4589-2, ASTM D2863, GB/T 2406.2, and JIS K 7201-2. These standards define the apparatus, specimen geometry, gas flow conditions, ignition procedure, and pass/fail criteria used to determine the LOI value. The test is highly reproducible, requires only small specimens, and produces a single quantitative figure, which makes it particularly valuable for quality control, flame-retardant formulation development, and comparative material screening. It should be noted, however, that the LOI is a relative ranking parameter obtained under controlled laboratory conditions and does not by itself constitute a full assessment of fire hazard in end-use scenarios.

LOI values are typically classified as follows:

• LOI below 21%: the material burns readily in normal air and is considered easily combustible.
• LOI between 21% and 27%: the material burns slowly in air and may be described as slowly burning.
• LOI above 27%: the material is difficult to ignite in air and is generally classified as flame retardant or self-extinguishing.

Testing Samples

Rigid Self-Supporting Plastics

For standard rigid materials such as PE, PP, ABS, PS, PVC, PA, PC, epoxy, and phenolic molding compounds, the preferred specimen is 80 mm to 150 mm long, 10 mm ± 0.5 mm wide, and 4 mm ± 0.25 mm thick. The specimen is clamped vertically in a stainless-steel holder that supports the lower portion while leaving the upper part free for ignition. Specimens should be smooth, free of surface defects, and representative of the material as supplied.

Flexible Plastics and Materials That Deform on Heating

Thin or flexible specimens that shrink, curl, or sag when heated cannot be tested unsupported. ISO 4589-2 defines alternative specimen forms and support devices, including U-shaped or framed holders, so that the specimen remains upright and the geometry of the test is preserved. The specific specimen type used must be stated in the test report, because LOI values depend strongly on specimen form and thickness.

Plastic Films and Thin Sheets

Very thin films and foils are tested using special preparation techniques described in the annexes of the relevant standards, typically by rolling the film onto a glass or metal rod to form a composite specimen that burns in a controlled manner. This approach prevents erratic flame behavior caused by the extremely high surface-to-mass ratio of unsupported films.

Cellular Plastics and Foams

Rigid and flexible cellular materials are tested using larger cross-sections, commonly approximately 10 mm × 10 mm, sometimes with a support frame to prevent collapse during the test. The cell structure and density of foams significantly influence their LOI, so specimens must be cut so that the cell structure is not damaged or compressed.

Specimen Quantity and Conditioning

A minimum of 15 specimens is recommended for a single LOI determination, and 20 or more specimens are advisable for unknown materials, because the bracketing procedure may require a larger number of trials. Unless otherwise specified, specimens shall be conditioned at 23 °C ± 2 °C and 50% ± 5% relative humidity in accordance with ISO 291 for at least 40 hours before testing. Moisture-sensitive materials such as polyamides may require additional drying, and cellulosic or hydrophilic materials must be tested promptly after conditioning since absorbed moisture can measurably raise the apparent LOI.

Testing Items

The primary deliverable of the test is the LOI value, expressed as volume percent oxygen to the nearest 0.1%. A complete test also records the following items, which are required by the standards for a full report:

• Complete identification of the material, including grade, formulation, color, and manufacturer's lot reference.
• Specimen dimensions, specimen type, thickness, and orientation relative to any anisotropy.
• Conditioning method, duration, and atmospheric parameters.
• Ignition method used, namely top-surface ignition or propagating ignition.
• Individual oxygen concentration, afterflame time, and extent of burning for each specimen.
• Observed burning behavior, including dripping, charring, glowing combustion, or smoky burning.
• The final LOI value and, where required, the estimated precision of the determination.

For certain applications, supplementary determinations may be performed, such as the temperature index according to ISO 4589-3, which measures LOI at elevated temperatures and provides additional insight into fire performance under realistic service conditions.

Testing Methods

Apparatus and Calibration

The standard LOI apparatus consists of a heat-resistant borosilicate glass chimney at least 450 mm high with an internal diameter of 75 mm to 100 mm, a vertical specimen holder, a small impinging flame igniter fueled by propane or butane, and a precision gas mixing and metering system supplied with oxygen and nitrogen. The total gas flow is regulated to produce a linear upward velocity of 40 mm/s ± 10% through the chimney, which for a 75 mm diameter chimney corresponds to approximately 10 L/min to 11 L/min. Flow measurement devices, such as calibrated rotameters or electronic mass-flow controllers, must be accurate to within ±0.5% of the indicated oxygen concentration, and the complete apparatus should be periodically verified using certified reference materials such as standardized PMMA panels with a known LOI.

Test Procedure

The determination follows a systematic bracketing procedure as specified in ISO 4589-2, ASTM D2863, and equivalent national standards:

• Step 1: Verify gas supplies, calibrate flow controllers, and allow the chimney atmosphere to stabilize. Mount the specimen vertically in the holder with its upper end positioned at the correct height inside the chimney.
• Step 2: Select an initial oxygen concentration based on experience with similar materials, typically 18% for expectedly flammable plastics or a higher value for flame-retardant compounds. Purge the chimney with the mixed gas for at least 30 seconds before ignition.
• Step 3: Apply the ignition flame to the specimen. For top-surface ignition, the flame is moved across the entire top of the specimen for up to 5 seconds; for propagating ignition, the flame is applied so that the top approximately 6 mm of the vertical faces is covered. Ignition attempts are repeated as necessary, but the total flame application shall not exceed 15 seconds.
• Step 4: Immediately after removing the igniter, start the timer and observe the burning behavior. A specimen is considered to have supported combustion if it burns for more than 180 seconds after removal of the flame, or if the flame front passes a reference mark 50 mm below the top of the specimen.
• Step 5: If the specimen fails, that is, burns beyond the criteria, reduce the oxygen concentration for the next specimen; if the specimen passes, increase the oxygen concentration. Initially the concentration is changed in increments of about 0.5%, and after the transition point has been bracketed, the increment is reduced to 0.2% or less to refine the determination.
• Step 6: Continue testing until the oxygen concentration is bracketed within the required resolution, then compute the LOI from the final balanced concentration or by linear interpolation of the last series of results.

Calculation of the LOI

The limiting oxygen index is calculated from the volume flow rates of oxygen and nitrogen at the balanced condition as follows: LOI = 100 × V(O2) / [V(O2) + V(N2)], where V(O2) and V(N2) are the volumetric flow rates of oxygen and nitrogen respectively. The result is reported as a percentage to one decimal place. The report must also state the specimen type and thickness, ignition method, gas flow velocity, conditioning, and observed burning phenomena, because these parameters materially affect the numerical result.

Factors Affecting Results

• Specimen thickness: thicker specimens generally exhibit higher LOI values for many thermoplastics, so results are only comparable at equal thickness.
• Temperature: LOI decreases as ambient or specimen temperature rises, which is why tests are conducted at 23 °C ± 2 °C.
• Humidity and conditioning: absorbed moisture can suppress combustion and inflate the LOI of hygroscopic materials.
• Gas purity and flow calibration: impure gases or miscalibrated flowmeters introduce systematic bias into the oxygen concentration.
• Operator technique: inconsistent ignition timing and flame position are common sources of interlaboratory variation.

Interpretation and Relation to Other Fire Tests

The LOI provides a convenient quantitative ranking of relative flammability, but it does not directly map onto regulatory classifications. For example, no exact conversion exists between LOI and UL 94 ratings, although many engineering thermoplastics that achieve a UL 94 V-0 classification also exhibit LOI values above approximately 30%. LOI data are therefore best used in combination with other fire performance tests, such as UL 94, IEC 60695 series tests, glow-wire tests, or cone calorimetry, to build a complete picture of a material's fire behavior.

Applications

LOI testing is applied across virtually every industry in which polymeric materials are used near potential ignition sources. Typical applications include:

• Electrical and electronic engineering: flammability screening of enclosure materials, connectors, switches, and printed circuit board laminates in support of UL 94 and IEC 60695 compliance programs.
• Wire and cable: evaluation of insulation and sheathing compounds for flame-retardant cable designs.
• Construction: supplementary characterization of panels, pipes, thermal insulation foams, and engineering plastics used in building products alongside EN 13501 and GB 8624 classification schemes.
• Transportation: material qualification for rail vehicle components under EN 45545 requirements, aircraft interior materials, and automotive interior parts.
• Flame-retardant research and development: rapid screening of halogenated, halogen-free, mineral-filled, and intumescent flame-retardant formulations during compound optimization.
• Quality control and incoming inspection: lot-to-lot verification of flame-retardant resin consistency by manufacturers and accredited third-party laboratories.
• Academic and regulatory research: standardized data generation for fire science studies and material certification dossiers.

FAQ

What is a good LOI value for a flame-retardant plastic?

As a general rule, materials with an LOI above 27% are considered flame retardant, since they will not sustain combustion in ordinary air. Many industry specifications, particularly in rail and aviation applications, require LOI values of 30% to 32% or higher, while standard engineering applications often target 28% or above.

What are the differences between ISO 4589-2, ASTM D2863, GB/T 2406.2, and JIS K 7201-2?

These standards are technically harmonized and are all derived from the same fundamental method. GB/T 2406.2 and JIS K 7201-2 are closely aligned with ISO 4589-2, while ASTM D2863 contains minor differences in chimney dimensions, flow verification, and data treatment. For regulatory submissions, the specific standard invoked by the customer or the target market must be used, and results obtained under different standards should be compared with caution.

How many specimens are required for one LOI determination?

A minimum of 15 specimens is normally sufficient, but up to 20 or more may be needed when the transition point is difficult to locate, for example with materials that exhibit glowing combustion, erratic flaming, or strong thickness sensitivity. Testing an inadequate number of specimens is a common cause of poor repeatability.

Does a higher LOI guarantee a better UL 94 rating?

No. The two tests measure different phenomena. The LOI measures sustained burning in a controlled oxygen-enriched atmosphere, whereas UL 94 evaluates ignition from a small flame in air, including afterflame time and flaming drips. A high LOI strongly suggests good self-extinguishing behavior, but UL 94 performance must be verified by conducting the UL 94 test itself.

Why do LOI results sometimes differ between laboratories?

Common sources of variation include differences in specimen thickness and type, conditioning humidity, gas purity, flow meter calibration, ignition technique, and the concentration step size used during bracketing. Close adherence to the specified procedure and periodic apparatus verification with reference materials minimize interlaboratory discrepancies.

Can the LOI be measured at elevated temperatures?

Yes. ISO 4589-3 extends the method to elevated temperatures, allowing the determination of the temperature at which the LOI falls to a specified value, known as the temperature index. This information is useful for applications where materials operate at sustained elevated temperatures, such as electrical equipment and cable systems.