Exploiting wood species and grain orientation to illuminate the onset of pyrolysis and ignition in deliberate combustion

• Experiments characterize ignition of wood relevant to biomass user health. • Experiments separate influence of conductivity from permeability of wood during ignition. • Ignition characterized by pyrolytic acceleration rather than specific mass flux. • Time to ignition is predicted well by heat flux, but not a priori knowledge of material properties. During wood combustion the conversion of solid material to pyrolysis gases causes the release of condensable tar in the form of particulate matter. This particulate matter has severe consequences for human health when inhaled. Reduction of PM release can occur through the transition from pyrolysis to flame. In this study, variation in natural wood properties on pyrolysis and deliberate ignition of thermally thick wood are applied to understand theories of ignition for the purpose of informing real-world design and practice. Three hardwoods (Maple, Red oak, and White oak) and a softwood (Scots pine) were selected for their differing permeability. Experiments with pyrolysis only and with piloted ignition were done under otherwise identical conditions. Acceleration in the pyrolytic mass loss rate occurred at the same time as ignition with a pilot for all woods and grain orientations. Different grain orientations were employed to isolate the influence of thermal conductivity versus permeability on pyrolysis and ignition times. Samples with the longitudinal end grain oriented toward external heat flux have a higher thermal conductivity in the direction of heat flux and shorter time to ignition and pyrolysis. The effect of altered thermal conductivity is in the same direction but has a lower magnitude than expectations from theory. Changes in grain orientation show measurable effect of wood permeability on pyrolysis and ignition times. Across wood types, times to pyrolysis or ignition do not scale with thermal inertia as expected. Experiments were conducted at 30 kW/m 2 and 40 kW/m 2 and we report ratios of time to pyrolysis (median 1.85) and time to ignition (median 2.18) for each wood type and orientation. Finally, we assess energy requirements to achieve deliberate ignition which indicates marginal energy demand between grain configurations compared to increasing heat flux. Overall, theoretical prediction of relative onset time explains behavior of individual woods in different configurations but is less useful at extrapolating between woods.