Analysis of Smoke Generation Characteristics of XPS (Extruded Polystyrene) Insulation Boards

Smoke generation characteristics are among the most decisive fire-safety parameters for extruded polystyrene (XPS) insulation boards, governing their acceptance in walls, roofs, basements, and facade systems across global markets. This article provides a technical analysis of how and why XPS produces smoke, the key measurable parameters, and the standardized methods—such as ISO 5659-2, ASTM E662, ISO 5660-1, EN 13823, and GB/T 8323—used by accredited third-party laboratories to quantify smoke density, smoke growth rate, and smoke toxicity.

Technical Overview

Extruded polystyrene (XPS) board is a rigid, closed-cell foam thermal insulation material produced by continuously extruding polystyrene resin mixed with blowing agents, flame retardants, nucleating agents, and colorants. Because of its low thermal conductivity, high compressive strength, low water absorption, and dimensional stability, XPS is widely applied in inverted roofs, floor slabs, perimeter and below-grade walls, cold-chain facilities, and composite insulation systems.

From a fire engineering perspective, XPS is a combustible thermoplastic. When exposed to radiant heat or flame, it softens at approximately 90–100 °C, shrinks away from the heat source, melts into a liquid pool, and frequently drips burning droplets that can trigger secondary ignition. The styrene repeating unit in the polymer backbone contains an aromatic ring, which gives polystyrene one of the highest soot yields among common building polymers. Consequently, XPS fires typically generate large volumes of dense, black, obscuring smoke rich in carbon monoxide, carbon dioxide, unburned styrene monomer, aromatic hydrocarbons, and—when brominated flame retardants are present—hydrogen bromide.

Analyzing the smoke generation characteristics of XPS therefore serves three purposes: quantifying optical density and its effect on escape visibility, quantifying smoke growth rates that influence flashover and suppression timelines, and characterizing toxic gas yields that determine tenability limits for occupants. These outputs feed directly into product classification under GB 8624, EN 13501-1, and UL 723/ASTM E84 regimes, as well as into performance-based fire safety engineering.

Testing Samples

General-Purpose XPS Boards

Standard test programs begin with commercial-grade XPS boards of typical densities from 28 to 45 kg/m³ and thicknesses from 20 to 100 mm. Both the extruded skin surface and the cut cell structure are evaluated separately, because the intact skin and the open cells of a cut face absorb radiation differently, which measurably shifts ignition time, melt behavior, and smoke release.

Flame-Retarded XPS Grades

Most XPS sold for building applications contains flame retardants, historically hexabromocyclododecane (HBCD, now phased out under the Stockholm Convention) and, increasingly, polymeric brominated copolymers or phosphorus–nitrogen intumescent systems. Comparative smoke testing of retarded versus non-retarded formulations is essential, because halogenated systems often improve ignition resistance while increasing soot and carbon monoxide yields under ventilation-limited conditions.

Faced and Composite Products

Foil-faced boards, glass-mat or mortar-laminated panels, and boards embedded in external thermal insulation composite systems (ETICS) are tested in their end-use configuration. Facings can dramatically reduce smoke by shielding the foam from direct radiation, retarding melt collapse, and promoting char-like barrier formation, so the as-installed build-up must be represented in the specimen.

Specimen Preparation and Conditioning

Specimens are precision-cut to the dimensions required by each method, wrapped in aluminum foil on all faces except the exposed surface, and conditioned to constant mass at 23 ± 2 °C and 50 ± 5 % relative humidity in accordance with ISO 291 or the conditioning clauses of the relevant test standard. Thickness is reported because it influences melt pooling, sustained flaming, and cumulative smoke production. Sample identity, production date, retardant system, and aging history are documented, since retardant depletion through thermal or UV aging can increase smoke output over service life.

Testing Items

A comprehensive smoke characterization of XPS typically reports the following parameters:

• Light transmittance through the smoke-filled chamber as a function of time, and the derived specific optical density curve Ds(t).
• Maximum specific optical density Ds(max), a headline indicator of peak obscuration.
• VOF4, the cumulative specific optical density over the first four minutes, which reflects early smoke growth critical to evacuation.
• Smoke Density Rating (SDR) determined under GB/T 8627, used within the GB 8624 classification framework, where certain B1 classifications require SDR ≤ 75.
• Smoke Developed Index (SDI) from tunnel testing to ASTM E84/UL 723, reported on the 0–450+ scale used by North American codes.
• Smoke Production Rate (SPR) and Total Smoke Production (TSP) from cone calorimeter testing, together with Specific Extinction Area (SEA) in m²/kg, which normalizes smoke to specimen mass loss.
• Smoke Growth Rate Index (SMOGRA) from the Single Burning Item test, used to assign the s1, s2, or s3 smoke classes of EN 13501-1.
• Ignition time, flaming duration, mass loss rate, and melt/drip observations, which explain the physical mechanism behind the smoke data.
• Toxic gas concentrations and yields—CO, CO₂, styrene, benzene, HBr, HCl, HF, and HCN where applicable—together with Fractional Effective Dose (FED) tenability calculations.

Testing Methods

Smoke Density Chamber — ISO 5659-2, GB/T 8323.2, ASTM E662

The single-chamber smoke density test is the cornerstone method for XPS smoke characterization. A 75 mm × 75 mm specimen, up to 25 mm thick and foil-wrapped, is exposed horizontally to a radiant furnace set at 25 kW/m² in two modes: with a pilot flame (flaming mode) and without (non-flaming mode). A vertical photometric system continuously measures light transmission across the 0.51 m³ chamber, and specific optical density is calculated from the transmittance loss, chamber volume, exposed area, and optical path length. Marine applications under the FTP Code add a 50 kW/m² exposure and combine the chamber with FTIR gas analysis. For thermoplastic XPS, the flaming mode usually dominates, driven by melt pool combustion of the collapsed specimen.

Cone Calorimeter — ISO 5660-1, ASTM E1354, GB/T 16172

The cone calorimeter provides the richest dataset for fire modeling. A 100 mm × 100 mm specimen is irradiated at 25, 35, or 50 kW/m² under a controlled exhaust hood, where a laser photometer measures smoke extinction in the duct. The method yields smoke production rate curves, total smoke production, specific extinction area, heat release rate, and mass loss simultaneously, allowing smoke yield to be correlated with combustion intensity. Because gas flow is well characterized, cone data scale more reliably to real fires than static chamber data and are preferred inputs for zone and CFD evacuation models.

Single Burning Item Test — EN 13823

For the European Euroclass system, SBI testing of two wing specimens exposed to a 30 kW sandbox burner measures SMOGRA, the derivative-based smoke growth index. XPS boards are typically tested on a substrate representing the intended application. SMOGRA ≤ 30 m²/s² qualifies for class s1, values up to 180 m²/s² for s2, and higher values fall to s3; molten-collapse behavior of un-faced XPS frequently pushes results into the s2–s3 range, which is why facings or fire barriers are common in compliant systems.

Tunnel Test — ASTM E84 / UL 723

In the Steiner tunnel, a full-scale specimen is exposed to a controlled flame spread scenario and the Smoke Developed Index is derived from photometer readings in the exhaust duct. Foam plastics such as XPS are tested with specified facings and mounting, and code compliance commonly requires an SDI not exceeding 450 alongside the flame spread classification.

Smoke Toxicity Assessment — ISO 19700, ISO 19701, GB/T 20285

Dynamic tube furnace decomposition under ISO 19700, with gas analysis methods from ISO 19701 or FTIR coupled to the smoke chamber or cone, quantifies toxic species yields across a range of equivalence ratios. GB/T 20285 provides the toxicity grading referenced by Chinese classification practice. For XPS, the analysis focuses on CO asphyxiant potential, irritant styrene and aromatic species, and HBr where brominated retardants are used, since the polymer contains no nitrogen and does not yield HCN.

Applications

Smoke generation data for XPS boards directly support the following engineering and regulatory uses:

• Compliance demonstration for GB 8624 classification, where smoke density rating results accompany combustion performance grading for Chinese building approval.
• Achievement of Euroclass s1/s2 smoke classifications under EN 13501-1 for CE-marked facade and construction products.
• North American interior finish and foam plastic compliance using ASTM E84/UL 723 flame spread and smoke developed indices.
• Marine and offshore qualification under FTP Code requirements, combining chamber smoke density with gas toxicity limits.
• Product development, comparing flame-retardant and smoke-suppressant formulations, facing systems, and skin structures to reduce total smoke and smoke growth.
• Fire risk assessment and performance-based design, supplying TSP, SEA, and species-yield inputs for ASET/RSET tenability analysis and evacuation simulation of facades, tunnels, basements, and cold-storage halls.
• Aging and quality control studies, verifying that retardant migration and weathering do not degrade smoke performance below the certified level over the declared service life.

FAQ

Why does XPS produce heavier smoke than many other insulation materials?

The aromatic benzene ring in the polystyrene molecular chain is intrinsically soot-forming, so incomplete combustion yields a high fraction of carbonaceous particles. In addition, XPS melts and forms burning pools that sustain large diffusion flames, compounding soot production and carrying unburned styrene vapors into the smoke plume.

What is the difference between flaming and non-flaming smoke test modes?

In the non-flaming mode the specimen decomposes and smolders under radiation alone, while in the flaming mode a pilot ignites the pyrolyzate. For XPS the flaming mode usually produces markedly higher optical density because molten polymer combustion is more complete in heat release but far more soot-productive; both modes are reported since real fires transition between them.

Which parameters matter most for regulatory acceptance?

This depends on the target market: SDR ≤ 75 for certain GB 8624 B1 grades, Ds(max) and VOF4 with gas limits for marine FTP compliance, SMOGRA for EN 13501-1 smoke classes, and SDI ≤ 450 for typical North American code paths. A well-designed test program maps the required regime before testing begins.

How can smoke generation of XPS be reduced?

Effective measures include brominated polymeric or phosphorus–nitrogen intumescent retardants tuned for smoke suppression, metallic foil or mineral facings that shield the foam and stabilize it against melt collapse, laminated board constructions, and system-level fire barriers. Each countermeasure should be verified by comparative chamber and cone testing rather than assumed.

Does specimen thickness affect smoke results?

Yes. Thicker specimens provide more fuel for sustained melt-pool burning, increasing cumulative smoke measures such as TSP and VOF4, even though peak optical density may be governed more by the exposed surface. For this reason, thickness and mounting configuration must always accompany reported smoke data.

Can smoke data be used in evacuation modeling?

Cone calorimeter values—smoke production rate, specific extinction area, and species yields—are the preferred inputs for visibility and tenability calculations in zone and CFD models, because they are measured under quantified ventilation and irradiance and can be scaled to predicted fire scenarios with defined uncertainty.