The mechanism of flame spreading over the surface of igniting condensed-phase materials

Abstract

This paper describes an experimental and theoretical investigation of the fundamental mechanism by which a flame spreads over the surface of a condensed-phase materials in a quiescent gaseous environment containing a component with which it can react chemically. It is postulated that the advancing flame vaporizes the surface material lying before it. As these vapors diffuse away from the surface, they undergo an exothermic reaction with the chemically active component in the gaseous environment, and ignite; thus, flame spreading is viewed as continuous diffusive, gas-phase ignition. Flame-spreading velocities have been measured for a variety of solid materials in O2/inert environments between 4 and 415 psia. Well-defined experimental, conditions yielded reproducible results, and thus suggest that flame-spreading, velocity is an intrinsic combustion quantity. All data can be correlated by a power-law relationship between the flame-spreading velocity (V) and two gas-phase parameters-pressure (P) and reactive component mole fraction (Yox)—in the form V ∞ ( P Y m ) Φ . It is concluded that V is controlled by a gas-phase physical process—probably either heat or mass transfer—which supports the mechanism proposed. Temperature distributions ahead of the propagating flame were obtained from surface-mounted, fine-wire thermocouples. The temperature level as the flame passes over the thermo-couple bead is independent of P, Yox, and inert diluent, and about 120°C below that measured previously during steady-state vaporization. Thus, it is concluded that direct surface attack by oxygen is unimportant during flame spreading and that the transient vaporization phenomenon is probably quite different than that of steady pyrolysis. The mathematical statement of the postulated flame-spreading mechanism is sufficiently complex that a complete analytical solution is currently impossible. Postponing numerical solutions, simplistic analyses were conducted that resulted in predicted flame-spreading characteristics that were well supported by the data obtained over the entire range of experimentation. Based on the evidence presented, the authors conclude that the postulated theory is probably valid, and engineering design of systems involving flame-spread control now can be put on a rational basis.