Experimental Validation of a Solid-Phase Model for Wood Ignition and Burning

Abstract

An unsteady one-dimensional model for the solid phase is applied to simulate the spontaneous ignition and burning of thick wood samples with grain orientation parallel or perpendicular to incident heat fluxes in the range 18–40 kW/m2, that is, in the absence of flame. The description of heat and mass transfer processes, at constant gas pressure, is combined with global volumetric rates of wood decomposition and char oxidation. Surface regression occurs for a limit value of char density while a critical surface temperature describes ignition. Good quantitative predictions are obtained for the ignition times and the surface temperature and mass loss rate profiles during burning. Conversion always consists of three main stages. The first short transients correspond to the formation of a relatively thin charred surface layer and glowing ignition. The second much longer stage represents a pseudo-steady-state burning, where the rates of advancement of the decomposition and oxidation zones are approximately constant as well as the surface temperature and the global rate of mass loss. Finally, the last short stage, resulting from the adiabatic bottom condition, is the enlargement of the decomposition zone to the entire sample. The first transient stage and the ignition times are noticeably affected by the external heating conditions whereas the characteristics of the pseudo-steady-state burning are mainly dependent on the char oxidation rate, which is controlled by the oxygen diffusion rate.

Influence of gas and solid phase thermo-physical and transport properties on the thermo-chemical conversion of char in reacting media: intra-particle, microscopic and temporal mass loss-based sensitivity analysis

“…The λ correlation used by [5] and c p correlation used by [2] yield a deviation of 252% (at 1300 K) and 2976% (at 1400 K) respectively. (c) Major deviations are prevalent for other species as well, 800% (at 300K) for λ O2 using the correlation by [4], 100% for c p,CO using correlation by [6]. (d) In general, the estimates presented in Figure 1 match well in a small and discrete range of temperature but deviate significantly elsewhere.…”

“…Another sensitive aspect requiring due attention is the use of LS correlations independent of composition [6,7,9,10], resulting in deviations that remains highly sensitive to constituent specie concentrations. This aspect is particularly true in case of mixtures containing CH 4 and H 2 as these species have properties significantly different, an order of magnitude higher, as compared to other species (CO 2 , H 2 O, CO and N 2 all of which have relatively similar thermo-physical and transport properties).…”

“…The specie diffusivity in mixture, is approximated to be equivalent to (i) A constant [10] (ii) Equivalent to binary diffusivities (a) A constant [11] (b) A function of temperature [2,6,9,11,12] (iii) Approximated from unity Lewis number assumption [8,[13][14][15] The consequence of first and second considerations are described in reference to Figure 3 while aspects related to unity Lewis number are described in the next section. Figure 3 presents the variation of diffusivity as a function of temperature for species like O 2 , CO 2 , CO, H 2 O, N 2 and CH 4 in syngas mixture (box pattern).…”

“…The continuous line represents the diffusivity of H 2 in the syngas mixture estimated from the standard correlations. The temperature-dependent diffusivity and the constants as reported in the literature [6,10,16] are presented in the bandwidth (diamond pattern). It is observed that the diffusivity estimated by LS correlations deviate significantly, to the extent of an order of magnitude, in comparision to the estimates from standard correlations.…”

Simulating Thermal Wood Particle Conversion: Ash-Layer Modeling and Parametric Studies

“…Different models for wood combustion are available, with most of them being one-dimensional (1D). − Such 1D models allow for fundamental studies of combustion-related processes while, at the same time, remaining numerically efficient.…”