Numerical Simulations of Turbulent Mixing Layer Diffusion Flames

Turbulent diffusion flames are relevant in many areas of engineering, particularly in the power generation sector. In the limit of fast chemistry, the combustion is dictated by the mixing of chemical reactants within the mixing layer, a process which is dictated by the large turbulent structures present in the flow. After 70 years of research, the turbulent mixing layer is still not fully understood, which has a major impact on our ability to predict combustion efficiencies and pollutant formation in flames. In particular, the physical phenomena that drives the dependency of the flame temperature rise on the upstream flow conditions are unknown. Solving this problem is essential to improving our ability to predict turbulent flames.

The proposed project aims to investigate the role of turbulent structures in the combustion process of diffusion flames. This will be achieved through high-fidelity, time dependent numerical simulations of flames using a well-established simulation code. Extensive experimental data for H2/F2 diffusion flames is available in the literature, and will be used as a benchmark for validation of the simulation study.

The effect of the upstream flow conditions on the diffusion flame will be systematically assessed through varying the state of the boundary layers from which the mixing layer forms. The power of numerical simulation will allow careful control of the initial conditions of the flow, permitting a study of the factors that influence the flame development. The so-called Reynolds Number Effect, where the flame structure varies with increasing flow velocity, is not fully understood. The mechanisms that drive this effect will be studied in detail, and new relationships to describe this effect will be derived.

Diffusion flames of practical interest typically develop from initially turbulent conditions. The initially-turbulent mixing layer is known to behave in a markedly different manner to its initially-laminar counterpart. The physical processes behind this change in flow dynamics are not understood. This project will perform simulations of both non-reacting, and exothermically-reacting initially-turbulent mixing layers in order to provide insights into this problem.

A further complication of flows undergoing exothermic reaction is that the heat release serves to modify the flow dynamics. This project will study the effect of heat release on the flow structure through a systematic variation of the concentration ratio of the reactants.

The project will place a heavy demand on High Performance Computing. ALICE2 will be used extensively for the simulations. Use will also be made of the HPC Midlands service, a cross-institution supercomputing platform. The data produced will be published in internationally-leading journals, and presented at major conferences. The student will be expected to participate in these activities during the period of study.