Éric GUILLAUME, Damien Flammier, Per Blomqvist, Anna Sandinge, Thomas Rogaume, Jocelyn Luche, Talal Fateh, Bauke Knottnerus
ABSTRACT The quantitative analysis of fire effluents is essential for assessing both toxic hazards and combustion chemistry. ISO/TS 21397 specifies the use of Fourier transform infrared (FTIR) spectroscopy coupled to cone calorimeter tests for the online measurement of fire gases. To evaluate the robustness of this method, an interlaboratory round robin was conducted under the framework of ISO/TC 92. Six laboratories participated, four of which provided complete FTIR datasets. Four polymers were selected to represent different effluent profiles: high‐density polyethylene (HDPE), polymethyl methacrylate (PMMA), polyamide 6.6 (PA 6.6), and unplasticised polyvinyl chloride (PVC). Each laboratory performed three replicate cone calorimeter tests at 50 kW/m 2 . Combustion parameters (peak heat release rate, total heat release, effective heat of combustion) and effluent species were analysed. Qualitative results showed consistent identification of expected major species: carbon dioxide (CO 2 ) and carbon monoxide (CO) for all polymers, nitric oxide (NO) for PA 6.6 and hydrogen chloride (HCl) for PVC. Quantitative analysis focused on maximum concentrations and yields of CO 2 , CO, NO and HCl. Statistical treatment following ISO 5725 provided complete repeatability and reproducibility values. The results revealed a clear hierarchy in reproducibility: combustion parameters showed the lowest variability, followed by CO 2 , then CO, with NO and HCl exhibiting higher dispersion. Minor species such as HCN and NO 2 were detected inconsistently and excluded from statistical treatment. The study confirms that ISO/TS 21397 yields reproducible data for the major effluents of interest, while highlighting the need for more prescriptive guidance on calibration and quantification methods to further improve interlaboratory agreement.