Limiting Oxygen Index (LOI) Flame Retardancy Testing for Coatings

Limiting Oxygen Index (LOI) testing is a widely recognized method for evaluating the flame retardancy of coatings by measuring the minimum oxygen concentration required to sustain combustion. This article provides a comprehensive technical overview of LOI testing for coating materials, covering applicable standards such as ISO 4589-2, ASTM D2863, and GB/T 2406.2, sample preparation requirements, testing procedures, result interpretation, and typical industrial applications.

Technical Overview

The Limiting Oxygen Index (LOI), also known as the oxygen index (OI), is defined as the minimum concentration of oxygen, expressed as a volume percentage, in a flowing mixture of oxygen and nitrogen that will just support the flaming combustion of a material. The test is performed under controlled laboratory conditions using a vertical glass chimney in which a vertically mounted specimen is ignited from the top and burns downward, like a candle, in a precisely regulated oxygen-nitrogen atmosphere.

The LOI value is calculated as follows:

LOI (%) = [O2 / (O2 + N2)] × 100

Since ambient air contains approximately 20.9% oxygen, a material with an LOI below 21% will generally continue to burn freely in normal air, while a material with an LOI significantly above 21% will tend to self-extinguish. For coating materials, a higher LOI value indicates stronger flame-retardant performance. The LOI method is particularly valued for its good reproducibility, quantitative output, and relatively small sample requirements, making it suitable for both research and development and quality control of fire-retardant and intumescent coating formulations.

The principal standards governing LOI testing for coatings and polymeric materials include:

• ISO 4589-2 / EN ISO 4589-2 – Plastics – Determination of burning behaviour by oxygen index – Part 2: Ambient-temperature test
• ASTM D2863 – Standard Test Method for Measuring the Minimum Oxygen Concentration to Support Candle-Like Combustion of Plastics (Oxygen Index)
• GB/T 2406.2 – Chinese national standard, technically aligned with ISO 4589-2
• JIS K 7201-2 – Japanese industrial standard for oxygen index determination
• NES 714 – Naval engineering standard often referenced for fire performance of materials in marine applications

Testing Samples

Because coatings are typically supplied as liquids or powders, appropriate specimen preparation is essential to obtain meaningful LOI results. Accredited third-party laboratories generally accept the following sample forms:

Free Coating Films

Liquid coatings are cast onto release substrates such as polytetrafluoroethylene (PTFE) sheets, glass, or silicone-coated liners and allowed to cure fully under specified conditions. The resulting free film is then cut into rectangular strips, typically 80–150 mm in length and approximately 10 mm in width, with a thickness commonly in the range of 1–3 mm. Multiple layers may be laminated or cast to achieve the required thickness, provided that delamination does not occur during testing.

Coated Substrates

When the end-use configuration is of interest, coatings may be applied to their intended substrates, such as steel panels, wood boards, cable sheathing materials, or textile fabrics. In such cases, the reported LOI reflects the performance of the coating-substrate system rather than the coating alone, and this must be clearly stated in the test report.

Powder Coatings and Meltable Materials

Thermosetting powder coatings and hot-melt materials are compression-molded or melt-cast into plaques of the required thickness, which are subsequently machined or die-cut into standard test specimens.

Specimen Conditioning

Unless otherwise specified, specimens should be conditioned to equilibrium at 23 ± 2°C and 50 ± 5% relative humidity for a minimum of 88 hours prior to testing, in accordance with the conditioning requirements of ISO 291 or ASTM D618. A minimum of 15–20 specimens is normally required to complete a full LOI determination using the bracketing procedure.

Testing Items

Typical test items evaluated under a coating LOI test program include:

• LOI value determination – the primary quantitative result, reported as the volume percentage of oxygen at the critical burning point
• Self-extinguishing behavior assessment – verification of whether the coated material continues to burn or extinguishes in normal atmospheric air (21% oxygen)
• Comparative flame-retardancy ranking – evaluation of baseline coatings versus flame-retardant modified formulations, or before-and-after aging exposure
• Verification of flame-retardant grade claims – confirmation that a product meets a specified LOI threshold required by a customer specification or product standard
• Effect of film thickness – determination of the minimum dry film thickness (DFT) required to achieve the target oxygen index

Testing Methods

The LOI test is conducted using an oxygen index apparatus consisting of a heat-resistant glass chimney (typically 70–100 mm in diameter and 450–600 mm in height), a specimen holder, a gas mixing and flow control system with calibrated oxygen and nitrogen flowmeters, and a small propane or butane ignition burner. Testing is performed at ambient temperature (23 ± 2°C).

Test Procedure

• Mount the conditioned specimen vertically in the holder so that at least 50–60 mm of the specimen is exposed above the top of the chimney.
• Purge the chimney with the oxygen-nitrogen mixture at the specified total flow rate (typically 10–13 L/min per ISO 4589-2) for at least 30 seconds to establish a stable atmosphere.
• Apply the ignition flame to the top of the specimen for up to 30 seconds, withdrawing and reapplying the burner as necessary, until the specimen is ignited to a depth of at least 5 mm.
• Observe the combustion behavior. A specimen "passes" at a given concentration if it either extinguishes before burning or after glowing extends 50 mm down from the top, or if it extinguishes within 180 seconds of ignition-source removal, whichever occurs first; otherwise it "fails."
• Adjust the oxygen concentration upward or downward in defined increments and test fresh specimens until the critical concentration is bracketed.
• Refine the result using the specified incremental steps (e.g., 0.2% oxygen intervals) and, where required, apply the Dixon and Mood statistical method or the bracketing factor method to calculate the final LOI value.
• Report the LOI rounded to 0.1 or 0.2 percentage points, together with specimen dimensions, thickness, orientation, and any observed burning phenomena such as dripping, char formation, or intumescence.

Interpretation of Results

While classification thresholds vary by product standard and application, the following general guidance is commonly applied:

• LOI < 21% – readily flammable in normal air; the material will sustain combustion after ignition
• LOI 21–27% – self-extinguishing tendency in air; generally regarded as slow-burning
• LOI > 27% – considered flame-retardant; the material is difficult to ignite and will not sustain combustion in ambient air
• LOI > 32% – high flame retardancy, often required for specialty applications in rail, marine, and aviation sectors

For intumescent fire-protective coatings, the formation and stability of an expanded char layer during the test is often noted as supplementary qualitative evidence of protective performance.

Quality Assurance Considerations

To ensure accuracy and comparability, the test gas concentrations must be verified with a calibrated oxygen analyzer, the flowmeters must be traceably calibrated, and the operator must be qualified per the laboratory's accredited procedures. Interlaboratory reproducibility for LOI testing is typically on the order of ±0.5 to ±1.5 percentage points of oxygen, depending on the material.

Applications

LOI flame retardancy testing of coatings supports a broad range of engineering, regulatory, and commercial objectives:

• Fire-retardant coating development – screening of halogen-free, phosphorus-based, nitrogen-based, and intumescent additive packages during formulation optimization
• Steel structural fire protection – verification of thin-film and thick-film intumescent coatings used in buildings and industrial plants
• Wire and cable coatings – evaluation of flame-retardant varnishes, dip coatings, and outer sheath coatings for cable assemblies in power, telecommunications, and rail installations
• Wood and textile finishes – assessment of flame-retardant lacquers, impregnation coatings, and fabric back-coatings used in furniture, curtains, and interior decoration
• Railway and marine engineering – compliance demonstration for interior coating systems against fire safety specifications referencing EN 45545-2, IMO resolutions, and related frameworks that accept LOI as supporting evidence
• Electronics and electrical equipment – qualification of conformal coatings and insulating varnishes used on printed circuit boards and motor windings
• Quality control and supplier verification – routine batch testing of incoming flame-retardant coating raw materials and finished products
• Scientific research – academic and industrial studies on combustion mechanisms, synergy effects of flame retardants, and thermal degradation behavior

FAQ

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

A coating system with an LOI of 27% or higher is generally considered flame-retardant, as it will not sustain combustion in ordinary air containing about 21% oxygen. Many high-performance specifications for rail, marine, and electrical applications require LOI values of 30–32% or above. The exact requirement should always be confirmed against the applicable product standard or customer specification.

Can a liquid coating be tested directly?

No. The LOI method requires solid, self-supporting specimens. Liquid coatings must first be cast and fully cured into free films of the specified dimensions, or applied to a defined substrate and tested as a coating-substrate composite. Powder coatings are compression-molded into plaques before specimen cutting.

Which standard should I choose: ISO 4589-2, ASTM D2863, or GB/T 2406.2?

All three standards are technically similar and based on the same candle-like burning principle. ISO 4589-2 and GB/T 2406.2 are essentially harmonized and widely used for international and Chinese market submissions, while ASTM D2863 is predominantly referenced in North American specifications. The choice depends on the regulatory market, the purchasing specification, and the certification scheme involved.

How is LOI different from other fire tests such as UL 94 or the cone calorimeter?

The LOI test measures the ease of ignition and sustained burning in a controlled atmosphere and yields a single quantitative index. UL 94 evaluates the burning behavior of specimens oriented horizontally or vertically with a small flame under specific timing criteria, while the cone calorimeter measures heat release rate, smoke production, and mass loss under radiant heating. These methods are complementary; LOI is often used for screening and specification purposes, with the other tests providing performance data under more severe fire exposure conditions.

How many samples are needed for an LOI test?

A complete determination typically requires 15 to 20 specimens, because the bracketing procedure involves repeated trials at multiple oxygen concentrations. Specimens should be uniform in thickness and free from voids, bubbles, and edge defects, as these irregularities can significantly influence the result.

Does specimen thickness affect the LOI result?

Yes. LOI values generally increase with specimen thickness for many materials, so the tested thickness must always be reported along with the result. When comparing formulations or verifying a specification, specimens of identical thickness and orientation should be used.

How long does an LOI test take?

The test itself is relatively rapid, and a full determination on a well-prepared set of specimens, including conditioning, is usually completed within one to three working days at an accredited third-party laboratory, subject to laboratory scheduling and the complexity of the sample preparation required.