The Fire Hazards of Insulation Materials

McKenna, Sean Thomas (2019) The Fire Hazards of Insulation Materials. Doctoral thesis, University of Central Lancashire.

[thumbnail of Thesis Document]
Preview
PDF (Thesis Document) - Submitted Version
Available under License Creative Commons Attribution Non-commercial Share Alike.

34MB

Abstract

Insulation materials are widespread in the modern built environment. They have, particularly in recent years, been a major focus of fire safety research. That focus has been enhanced by the tragic Grenfell Tower Fire that resulted in the death of 72 people. This work aims to understand and quantitatively assess the fire hazards presented by modern insulation materials.
14 materials were selected for analysis, including 7 PIR foams, 4 phenolic foams and 3 mineral wool materials. These materials were tested for their elemental composition, fire toxicity, and reaction-to-fire properties. The data generated was then used to calculate the maximum safe loadings of the insulation materials. The methodology had originally only been used with estimated values based on Euroclass data. In order to practically apply the method, the cone calorimeter was used to generate the mass loss per unit area data, rather than SBI test data or estimated values. Fire toxicity data was generated using the ISO/TS 19700 Steady State Tube Furnace. Additional maximum safe loading values were calculated using material-IC50 values, as incapacitation is arguably a more important end point in fire toxicity assessment. The maximum safe loading values calculated were comparable to the estimated values outlined in the original methodology. This methodology could be used to provide quick estimations of the safe loading of insulation materials in construction, allowing for informed decision making in building design without an overwhelming amount of data for non-fire experts to consider.
The results of this work demonstrate significant differences between the 3 types of insulation material. The mineral wool materials (both glass wool and stone wool) were of low toxicity and flammability. The foam insulation materials (PIR and phenolic) produced high yields of toxic gases in under-ventilated conditions, and had relatively high flammability. The PIR foams, in particular, had the highest toxicity due to the high yields of HCN produced during under-ventilated flaming, which has been linked to their nitrogen content and chemical composition. The phenolic foams lacked the high yields of HCN due to their low nitrogen content, but still produced high quantities of asphyxiating CO, like the PIR foams. Both types of foam insulation also produced hydrogen chloride gas during combustion, which would have a strongly irritating effect on exposed persons, potentially hindering their escape. FED analysis has demonstrated that the PIR foams increased toxicity is largely the result of the high toxicity of HCN. 1 kg of any of the 7 PIR samples burning in under-ventilated conditions is capable of producing enough HCN to create a lethal atmosphere in 50 m3. The maximum safe loading values calculated showed that, on average, phenolic foams present ~50 to 100x higher fire hazard than the mineral wool materials, and the average PIR foam presented a potential fire hazard ~1.5 to 2.5x higher than the average phenolic foam.
Additional work was performed to optimise a method for the quantification of HCN in fires – the chloramine-T/isonicotinic acid method from ISO 19701. HCN is a highly toxic product of the combustion of nitrogen containing materials. As such, it was important to ensure sampling and analysis was both accurate and reliable. Analysis was performed to understand sample and standard stability, optimal time to analysis, analytical variation, and potential interferences as a result of commonly encountered acid gases in fire effluent.
The cone calorimeter and SBI apparatus were also assessed for their viability in fire toxicity assessment, potentially negating the need to use the ISO/TS 19700 Steady State Tube Furnace. However, the resulting data demonstrated that both tests are inadequate due to their inability to recreate the more toxic fire condition – under-ventilated burning. This emphasises the need for dedicated fire toxicity tests, as most fire tests are well-ventilated reaction-to-fire tests, despite the fact that fire toxicity results in at least 50% of UK fire deaths.


Repository Staff Only: item control page