Smoke composition using MLC/FTIR/ELPI: Application to flame retarded ethylene vinyl acetate Franck-Estime Ngohang a, b , Ga € elle Fontaine b , Laurent Gay a , Serge Bourbigot b, * a MFEE/EDF R&D, 6 Quai Watier, 78400 Chatou, France b UMET/ENSCL, Avenue Dimitri Mendeleïev, CS 90108, 59652 Villeneuve d’Ascq, France article info Article history: Received 19 December 2014 Received in revised form 26 February 2015 Accepted 1 March 2015 Available online 9 March 2015 Keywords: Smoke Flame retardant Aluminum trihydroxide EVA FTIR ELPI abstract This paper is devoted to the interpretation of smoke data in terms of the fire behaviour of ethylene vinyl acetate unfilled and filled with aluminium trihydroxide (EVA and EVA/ATH), under different fire sce- narios: 25, 35, 50 and 75 kW/m 2 . Flammability parameters, evolved gases and soot particles were evaluated for both materials (EVA and EVA/ATH) as a function of applied heat flux, using a combined Mass Loss Cone, Fourier Transform Infrared spectroscopy and Electrical Low Pressure Impactor (MLC/ FTIR/ELPI) in a simultaneous analysis bench test. It was found that the EVA formulation leads to the formation of a carbonaceous layer (char) at low heat flux (25 kW/m 2 ), which blocks the release of decomposition gases, including flammable gases, thereby delaying the onset of piloted ignition of EVA relative to EVA/ATH. Total Heat Release (THR) measurements during tests performed on EVA show that this value remains constant at all heat fluxes, which is consistent with the fact that no residue of the EVA samples remains at the end of the fire tests. In the case of EVA/ATH however, the measured THR is proportional to the external heat flux because the quantity of combusted material increases as a function of the thermal stress applied. Calculation of the average effective heat of combustion (AEHC) of the EVA and EVA/ATH material yielded similar values, in both cases, under all four fire scenarios. With respect to gas phase analysis, the concentration of acetic acid (CH 3 COOH) release in the smoke of EVA (from the de- acetylation of vinyl acetate) was found to increase as the heat flux decreases, with a significant difference at 25 kW/m 2 relative to 35, 50 and 75 kW/m 2 . Concerning EVA/ATH, it appears that there is a trans- formation of CH 3 COOH to acetone (CH 3 COCH 3 ) attributed to the catalytic effect of Al 2 O 3 , which increases as the heat flux increases. Calculation of gas species yields revealed that the CO yields of EVA/ATH are higher than those of neat EVA, which can be ascribed to the incomplete combustion reaction of the flame retarded material (EAV/ATH) as compared to neat EVA; an exception was nevertheless noted at 25 kW/ m 2 , where carbonization of EVA was also reported. In relation to ex situ analysis of particle size distri- bution and concentration in the EVA and EVA/ATH smoke, it was shown that ultrafine and fine particles are present in much higher concentration than particles above 1 mm(1 mm) for both formulations at all applied heat fluxes. Nonetheless, the more ultrafine particles (those of 6 nm) were solely detected in the EVA/ATH smoke, suggesting that the ATH flame retardant additive may promote the release of these ultrafine particles. © 2015 Elsevier Ltd. All rights reserved. 1. Introduction Investigations into the fire behaviour of polymeric materials are often conducted at 35 and/or at 50 kW/m 2 [1,2], because these heat fluxes best reproduce the thermal aggression and heat stresses generated respectively by a mild fire and/or a flashover fire situation. When it comes to determining thermal stability or degradation of materials during flaming combustion, tests at different thermal stresses can lead to a better understanding of fire performance as well as improve knowledge of the transient evo- lution of gases in terms of decomposition processes and the type of soot particles emitted in the smoke. The focus of this paper is to investigate the effect of heat flux on the characteristic fire behav- iour of ethylene vinyl acetate with or without aluminium tri- hydroxide. Following a novel approach supported by the coupling of Mass Loss Cone, Fourier Transform Infrared Spectroscopy and * Corresponding author. Tel./fax: þ33 3 20 43 48 88. E-mail address: serge.bourbigot@ensc-lille.fr (S. Bourbigot). Contents lists available at ScienceDirect Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab http://dx.doi.org/10.1016/j.polymdegradstab.2015.03.002 0141-3910/© 2015 Elsevier Ltd. All rights reserved. Polymer Degradation and Stability 115 (2015) 89e109