Three-dimensional simple chemistry Direct Numerical Simulations (DNS) of Bunsen burner flames have been carried out for different pressure values. A number of cases have been considered for the same set of values of mean and root-mean-square inlet velocities normalised by the laminar burning velocity and the integral length scale normalised by the nozzle diameter. The modifications of laminar burning velocity and flame thickness with pressure lead to an increase in both flow and turbulent Reynolds numbers with increasing pressure. This also gives rise to changes in Damköhler number and Karlovitz numbers for these flames and thus they occupy different locations on the regime diagram. For this reason, two additional cases at the lowest pressure have been simulated to match the turbulent Reynolds number of the highest-pressure case by changing the normalised root-mean-square velocity in one case, whereas the integral length scale is modified in the other case. It has been found that pressure and turbulent Reynolds number variations do not have significant influences on the mean behaviours of the magnitude of the reaction progress gradient (i.e. Surface Density Function) and fluiddynamic normal strain rate. However, the length scale separation between the nozzle diameter and flame thickness increases with increasing pressure, which makes the occurrence of the Darrieus-Landau (DL) instability highly likely for the flames at elevated pressures. The presence of the DL instability affects the flame curvature statistics, which in turn influence the mean behaviours of the dilatation rate and fluid-dynamic tangential strain rate.

The validity of Damköhler's first hypothesis, which relates the turbulent flame speed to turbulent flame surface area under the condition where the integral length scale of turbulence is greater than the flame thickness, has been assessed using three-dimensional Direct Numerical Simulations (DNS) of turbulent premixed Bunsen burner flames over a range of values of Reynolds number, pressure and turbulence intensity. It has been found for the Bunsen configuration that the proportionality between volumeintegrated burning rate and the overall flame surface area is not strictly maintained according to Damköhler's first hypothesis. The discrepancy is found to originate physically from the local stretch rate dependence of displacement speed, and this helps to explain differences observed previously between flames with and without mean curvature. Approximating the local flame propagation speed with the unstrained laminar flame speed is shown to be inaccurate, and can have a significant influence on the prediction of the overall burning rate for flames with non-zero mean curvature. Using a twodimensional projection of the actual scalar gradient for flame area evaluation is shown to exacerbate the loss of proportionality between volume-integrated burning rate and the overall flame surface area.The current analysis identifies the conditions under which Damköhler's hypothesis remains valid and the necessary correction for non-zero mean flame curvature. Further, it has been demonstrated that surface-weighted stretch effects on displacement speed need to be accounted for in order to ensure the validity of Damköhler's hypothesis under all circumstances. Finally, it has been found that the volumeintegrated density-weighted scalar dissipation rate remains proportional to the overall burning rate for all flames considered here irrespective of the value of Reynolds number, pressure and turbulence intensity. However, this proportionality is lost when the scalar dissipation rate is evaluated using the two-dimensional projection of the actual scalar gradient.

Newcastle University ePrints -eprint.ncl.ac.uk Dopazo C, Cifuentes L, Alwazzan D, Chakraborty N. Influence of the Lewis number on effective strain rates in weakly turbulent premixed combustion. ABSTRACTThe influence of the global Lewis number, Le, on the statistical behaviour of the 'effective' normal and tangential strain rates have been analysed based on three-dimensional DNS data of freely propagating statistically planar turbulent premixed flames with Le = 0.34, 0.60, 0.80, 1.00 and 1.20. The volumetric dilatation rate is found to be mostly positive and its magnitude increases with decreasing Le. The flow normal strain rate predominantly assumes positive values and thus tends to pull adjacent iso-scalar surfaces apart, which reduces scalar gradients.By contrast, the "added" normal strain rate due to derivatives of the displacement speed normal to iso-surfaces has the propensity to push them closer together, and therefore increase the magnitude of scalar gradients. The balance between flow and added normal strain rates along with the advective transport determines whether scalar gradients are enhanced or destroyed.Iso-surface elementary area stretching by the fluid flow increases with decreasing Lewis number, and the added tangential strain rate exhibits predominantly negative values and is determined by the correlation between displacement speed components and flame curvature.It has been found that turbulent flames with small values of Lewis number exhibit flame thinning and high values of the flame surface area and these tendency strengthens with decreasing Lewis number. This behaviour has been explained in detail in terms of the statistical behaviours of effective normal and tangential strain rates.

No abstract

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Highlights  Surface Density Function transport analysed for DNS of turbulent Bunsen flames  Effects of pressure and Reynolds number variations analysed using DNS data  Effective normal strain rate remains insensitive to the pressure variation  DL instability at high pressure affects tangential strain and curvature stretch  Mean contributions of curvature stretch and tangential strain counter each other