Spherical Flame Front Thickness and Instability

Abstract

The spatial profile of a combustion wave has significant implications on flame stability, chemistry rates, and species diffusion. Although combustion science relies on flame thickness in characterizing flame propagation, consensus has not been achieved on the appropriate definition and measurement of flame thickness. In this work, emission profiles from excited methylidyne (CH*) and hydroxyl (OH*) were measured from spherically expanding flames for two hydrocarbon fuels (methane and propane) at three pressures (0.5, 1, and 2 atm) and a range of equivalence ratios (0.8 to 1.5). An Abel inversion algorithm was employed to account for the line-of-sight nature of the imaging data.

Species profiles predicted by AramcoMech 1.3 showed excellent agreement with the Abel-transformed experimental measurements, especially when accounting for image blurring experienced by the experiment. The effect of magnification was investigated by repeating experiments for four levels of imaging pixel density. Measured flame thicknesses were found to be highly dependent on pixel density, and this work demonstrates that 13.5 µm/pix is sufficient to resolve flame structures on the order of 200 µm with uncertainty of approximately 10%, even with burned-gas flame speeds on the order of 300 cm/s. The measurements indicate that the measured flame zone thickness based on electronically excited species is much closer to the length scale typically predicted by kinetics models than what has been seen in most experiments to date.

Laminar flame stability relies on multiple stabilizing and destabilizing phenomena, dependent on many properties of the gas mixture and the flame which largely must be predicted from thermodynamic and chemical kinetic modeling. Experimental critical-radius data were extracted from spherically expanding flames for 10-atm, stoichiometric syngas mixtures across a range of super-unity Lewis numbers (1.5 to 2.4). The onset of instability is shown to be strongly correlated with the Lewis number. The leading theoretical model predicts the qualitative experimental trends well, but significantly over- or under-predicts the critical radius, depending on the chosen formulation of flame thickness. Analysis shows that the hydrodynamic flame thickness could be directly related to the density gradient thickness, which produced the most qualitative agreement between modeled critical-radius results and the experiment.