Numerical and experimental studies of extinguishment of cup-burner flames by C₆F₁₂O

Abstract

The extinguishment of propane cup-burner flames by a halon-replacement fire-extinguishing agent C₆F₁₂O (Novec 1230) added to coflowing air in normal gravity has been studied computationally and experimentally. The time-dependent, axisymmetric numerical code with a detailed reaction mechanism (up to 141 species and 2206 reactions), molecular diffusive transport, and a radiation model, is used to reveal a unique two-zone flame structure. The peak reactivity spot (i.e., reaction kernel) at the flame base stabilizes a trailing diffusion flame, which is inclined inwardly by a buoyancy-induced entrainment flow. As the volume fraction of the agent in the coflow is increased gradually, the total heat release increases up to three times due to unwanted combustion enhancement by exothermic reactions to form HF and CF₂O in the two-zone trailing flame; whereas at the base, the flame-anchoring reaction kernel weakens (the local heat release rate decreases) and eventually the flame blows off. A numerical experiment, in which the C₆F₁₂O agent decomposition reactions are turned off, indicates that for addition of inert C₆F₁₂O, the maximum flame temperature decreases rapidly due to its large molar heat capacity, and the blow-off extinguishment occurs at ≈1700 K, a value identical to that for inert gases previously studied, while the reaction kernel is still burning vigorously. The calculated minimum extinguishing concentration of C₆F₁₂O in a propane flame is 4.12 % (with full chemistry), which nearly coincides with the measured value of 4.17 ± 0.30 %.

Introduction

Research for the last two decades to find a replacement for the effective but ozone-destroying fire suppressant Halon 1301 (bromotrifluromethane, CF₃Br) used in aircraft cargo-bays has made good progress; however, a replacement agent with all of the properties of CF₃Br has not yet been identified. Unlike CF₃Br, some replacement agents failed a mandated Federal Aviation Administration (FAA) Aerosol Can Explosion Test [1,2] as they caused overpressures due to unwanted combustion enhancement, as has been reported in the literature [3], [4], [5], [6], [7], [8], [9], [10]. Based on thermodynamic equilibrium and perfectly stirred-reactor calculations for premixed systems, it was postulated [11], [12], [13] that higher overpressures in the FAA aerosol can tests might be due to higher heat release from reaction of the inhibitor itself. Nonetheless, the agents should still reduce the overall reaction rate and inhibit the reaction [14,15] in these stirred-reactor systems since all reactants, intermediates, and products exist simultaneously. In diffusion flames and most fires, however, the flame structure, i.e., the distributions of the temperature, species concentrations, and velocity, play decisive roles in the interactions between the inhibitor and the flame. In turn, such interactions change the flame structure. Therefore, there is an urgent need to study the effects of a potential halon replacement agent on the chemical structure and extinguishment mechanisms of diffusion flames.

In previous papers [16], [17], [18], the results of full-chemistry computations of cup-burner flames with CF₃Br, C₂HF₅ (pentafluoroethane, HFC-125), C₂HF₃Cl₂ (2,2-dichloro-1,1,1-trifluoroethane, HCFC-123), and C₃H₂F₃Br (2-bromo-3,3,3-trifluoropropene, 2-BTP) added to the coflowing air were reported. Additional numbers of carbon and fluorine atoms in the halon-replacement-agent molecules, compared to CF₃Br, represent potential energy contributions at a fixed concentration if they burn completely to COF₂ and HF. The combustion enhancement was evident in the computation, particularly with C₂HF₅ or C₂HF₃Cl₂ added [18]. For C₂HF₃Cl₂ or C₃H₂F₃Br added, however, the calculation was unable to obtain the converged solution near the extinguishment limit. An attempt was unsuccessful even at low concentrations of one other agent C₆F₁₂O (dodecafluoro-2-methylpentan-3-one, FK-5–1–12, Novec 1230). C₆F₁₂O is a good agent on a mole basis but (like most other replacements) still quite inferior to CF₃Br on a mass basis. In fact, the mass-basis extinguishing concentration of C₆F₁₂O is comparable to chemically passive gases such as N₂, CO₂, CF₄, and C₄F₁₀, and C₆F₁₂O maybe acts like an inert fire-extinguishing agent. However, the physical and chemical effects of C₆F₁₂O on the diffusion flame structure and extinguishment processes have not yet been fully studied.

This paper extends the prior effort [18] on halon-replacement agents to C₆F₁₂O by overcoming the convergence difficulties in the numerical simulation and by conducting the cup-burner flame extinguishment experiments using propane as the fuel.