Fully Involved Enclosure Fires: Effects of Fuel Type, Fuel Area and Geometry

M. Delichatsios,Gordon Silcock
DOI: https://doi.org/10.3801/iafss.fss.7-59
2003-01-01
Fire Safety Science
Abstract:Flame spread and fire over cellulosic materials occur when the burning region supplies sufficient heat to the virgin solid to cause gasification. Under proper conditions, the reaction between the generated volatiles and the oxidant (air) may be sustained. The characteristics of the flame spread are therefore the result of heat and mass transfer processes as well as finite-rate gas phase chemical kinetics. In this paper, the development of the wood pyrolysis model and then the new three-dimensional mathematical model to study flame spread over timber wall lining are described. Both the pyrolysis and burning of a vertical timber wall of a room and the turbulent flow, combustion and radiation in the room, are included. The predictions of the pyrolysis and fire models have been compared with the results of experiments. Excellent agreement has been achieved. The prediction of flashover behaviour of enclosure fire is also discussed. INTRODUCTION Flame spread and fire over cellulosic materials such as wood are important issues in fire safety design due to the frequent involvement of such materials in building fires and to the severity of the fires. Experimental studies to understand the mechanism of flame spread have been reported . However, efforts to develop suitable mathematical models of flame spread in conditions of practical importance are not well developed. Many models, which are often classified as zone models, use empirical correlations derived from experiments. However the applications of these models are usually restricted to a particular set of fire scenarios and are, therefore, limited. Field models based on CFD techniques have been used to study the behaviour of enclosure fires. Applications of CFD models for fire safety studies have been reported . However, they did not take combustion into account but rather focused on a non-reacting gaseous flow in which the actual combustion processes were simulated by spatially distributed heat sources. Most importantly, these models did not include the pyrolysis of the solid which is the essential process governing the emission of combustible volatile pyrolysis products from the solid as the source of fuel. As a result, the emission of fuel gases from burning solids was not incorporated. In order to model the complex behaviour of flame spread over cellulosic materials, a comprehensive field model, is required. In this paper, we describe a new three-dimensional field model which includes the pyrolysis of the cellulosic material, emission of combustion gases from the material, gas phase combustion and feedback of heat from the burning region to the solid, causing further gasification. The model was used to predict the flame spread over cellulosic material such as timber wall lining and the subsequent fire development in enclosures until flashover. Excellent agreement was obtained. WOOD PYROLYSIS MODEL In the solid phase, multiple step first-order reactions are assumed for the pyrolysis of various active constituent of the wood becoming combustible gases. Evaporation of moisture in the wood is Copyright © International Association for Fire Safety Science 70 assumed to be sufficiently rapid for the vapour to be in equilibrium. Energy and mass transfer in the solid occurs by thermal conduction and convection due to the movement of vapour, volatile gases and inert gases through the porous wood material. Two conservation equations for energy and mass were formulated. When these equations are transformed into a non-orthogonal curvilinear coordinate system ξ to cater for possible complex geometries in enclosures, they can be written in the following generalised form: ( ) φ φ φ φ ∂ξ ∂φ β β φ β ∂ξ ∂ ∂ φ ρ ∂ S g F t g j j i i i = ⎟⎟ ⎠ ⎞ ⎜⎜ ⎝ ⎛ Γ − + l l l l l * [1] where φ denotes either temperature Ts or pressure Ps; superscript l = 1, 2 or 3 denotes the directions along and across the grains to account for the anisotropic properties of wood, and the quantities ρφ * , Γφ l , Fφl and Sφ are listed in Table 1. i l β , j l β and g represent the geometric coefficients and the Jacobian of the non-orthogonal curvilinear transformation. TABLE 1. Coefficients and source terms for variables in the solid phase conservation equations φ Ts Ps ρφ * ρ ρ s ps m pm C C + ( ) π g g s R T Γφ l λs l ( ) ( ) α ρ ρ ρ ρ ρ ρ s v m g g i i v g i M M M l + + + + Fφ l & & & m C m C m C g pg v pv i pi l l l + + 0 Sφ − − Δ Δ H R H R ev ev p p R M R M ev m p g + The quantities Cps, Cpm, ΔHev, and ΔHp are the specific heats of virgin wood and water, and heats of evaporation and pyrolysis, respectively. The subscripts g, v and i denote volatile gas, vapour and the inert gas which is assumed to be present initially in the wood at ambient conditions. In the wood pyrolysis model, the thermal decomposition of the cellulosic material and the evaporation of moisture are fully considered. The volatile gases and vapour are assumed to obey the ideal gas law with the gas and vapour densities assigned a dependence on porosity: The transfer of volatile gases and vapour is assumed to obey Darcy’s Law with anisotropic permeability dependent on the orientation of the grains. The thermal conductivity and volumetric heat capacity of the pyrolysing wood vary linearly with char fraction. The permeability of the pyrolysing solid was described by an exponential function of the varying solid density as the wood progressively transforms to char . All other thermodynamic and transport properties were allowed to vary linearly with temperature. Thermal swelling, shrinkage, surface regression and possible surface oxidation reactions of the virgin wood and char were neglected. The detailed development and formulation of the equations were reported in . NUMERICAL SIMULATION OF PYROLYSIS OF WOODEN CUBE IN A FURNACE To validate the 3-D wood pyrolysis model, it had been used to predict the pyrolysis of a wooden cube inside a furnace under various temperatures. The transformed governing equations were solved numerically using a control volume technique. Simulations were performed for the pyrolysis of wooden cubes of different initial moisture contents at furnace temperatures of 973 K and 1273 K. The activation energy and pre-exponential factor for the pyrolysis reaction of the single constituent have been adjusted to fit the measured mass loss history reported by Bonnefoy et al. 12
What problem does this paper attempt to address?