The localised forced ignition and early stages of flame development in a turbulent planar jet
Introduction
The localised forced ignition (spark, laser) of flammable mixtures is a topic of fundamental importance in combustion science [1]. Thanks to its simple geometrical configuration, localised forced ignition of jets has been experimentally extensively analysed in terms of ignition probability and flame propagation. Detailed experimental measurements of the ignition probability have been obtained by Ahmed and Mastorakos [2] for a methane turbulent round jet under different operating conditions, and focused mainly on the flame development subsequent to thermal runaway, and its subsequent upstream propagation and stabilisation, when applicable. They also demonstrated the importance of the edge flame propagation in the flame development [3], [4], [5] by measuring the flame displacement after ignition until its stabilisation. This experimental database has been extensively used to examine ignition models within the Large Eddy Simulation [6], [7], [8] and Reynolds-Averaged Navier–Stokes frameworks [9], [10]. However, to date, only a limited effort has been directed to the fundamental understanding of the flame evolution after localised forced ignition of inhomogeneous mixtures in shear-generated turbulence in which numerous flame regimes co-exist [11], [12], [13].
The present work aims to bridge this gap in the literature by discussing the extension of the current understanding of localised forced ignition gained from earlier Direct Numerical Simulation (DNS) analyses for homogeneous and inhomogeneous mixtures under isotropic decaying turbulence [14], [15], to shear-driven inhomogeneous flows. This is achieved by investigating the spark ignition of a turbulent planar methane-air jet using three-dimensional DNS with a two-step chemical mechanism. The study focuses on the understanding of the energy deposition location and initial flow field effects on the four stages of flame evolution, i.e. (i) kernel growth and advection, (ii) downstream and radial propagations, (iii) potential upstream flame propagation and/or (iv) flame stabilisation.
Section snippets
Mathematical formulation
A two-step chemical mechanism in which a pre-exponential adjustment is applied for rich mixtures has been used thanks to its low computational cost and ability to accurately estimate the laminar burning velocity across the whole flammability range [15], [16]. Furthermore, the addition of the reaction allows for a better prediction of the flow expansion around the flame [17]. The adequacy of the current mechanism to accurately reproduce the chemical effects is supported by
DNS database
The simulations have been carried out using the 3D DNS compressible code Senga+ [14]. The code employs a order central difference scheme for the internal grid points that transitions to a one-sided order scheme at the non-periodic boundaries for the spatial differentiation. The time advancement is carried out using a order explicit Runge–Kutta scheme.
The planar jet slot width is taken to be (where is the thermal flame thickness of the
Outcomes and general observations
For all the simulations, the ignition and propagation success (defined as the occurrence of a thermal runaway, and as the self-sustained growth of the kernel following external energy deposition respectively [14]) are recorded and summarised in Table 2.
No ignition was observed at location 2, irrespective of the realisation, thus confirming what could be surmised from the ignitor local conditions (see Table 1). Similar conclusions can be drawn for location 3, where a combination of high values
Conclusions
The localised forced ignition and subsequent flame propagation in a planar turbulent methane jet in ambient air has been simulated using Direct Numerical Simulation (DNS). Four different energy deposition locations have been selected, while for each of them, four independent realisations have been simulated using identical energy deposition parameters (width, duration, power).
The ignition/propagation success was found to be well correlated to the values of mean mixture fraction and flammability
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors are grateful to the British Council for financial support and to the EPSRC (EP/R029369/1, Archer, Cirrus) and Newcastle University (Rocket) for the computational support.
References (33)
Prog. Energy Combust. Sci.
(2009)- et al.
Combust. Flame
(2006) - et al.
Combust. Flame
(2016) - et al.
Proc. Combust. Inst.
(2017) - et al.
Combust. Flame
(2009) - et al.
Combust. Flame
(2015) - et al.
Combust. Flame
(2019) - et al.
Combust. Flame
(2019) - et al.
Combust. Flame
(1987) Combust. Flame
(1989)