1. What are the major characteristics of the temporal evolution of iron particles that heat up, ignite and burn in a turbulent flow?
2. How is the combustion process affected by the ambient gas conditions, particle cloud properties and turbulence characteristics?
3. What are the driving mechanisms of discrete/continuous iron dust flame stabilisation?

Left: Three-dimensional computational domain with gas velocity contour and iron particles coloured by their oxidation progress (black: non-oxidised, white: fully-oxidised). Right: Time evolution of gas temperature + particles coloured by their oxidation progress (top), oxygen mass fraction (middle) and flow vorticity (bottom) in the x-y plane at Lz/2.

A three-dimensional CP-DNS of a reacting dust cloud consisting of monodispersed iron particles in a turbulent mixing layer is conducted [1].

The upper stream of the mixing layer is initialised with cold iron particles with a uniform diameter of 10 micron suspended in air, while the lower stream consists of hot air flowing in the opposite direction. As time progresses, turbulent mixing entrains particles from the upper to the lower stream. These particles interact with the lower stream, heat up and begin their oxidation process. When the particles reach a critical temperature, they start to ignite and burn, as can be observed from the local decrease of oxygen and increase of gas temperature.

The number of computational cells for the CP-DNS is ~85 million, ~5 million particles are used and the computational cost corresponds to ~165,225 CPUh. This study shows that the major differences between non-volatile iron flames and volatile-containing solid fuel flames (e.g. coal/biomass) are non-vanishing particles at late simulation times and a stronger limiting effect of the local oxygen concentration on the overall conversion process in iron dust flames [1].

Representation of the time evolution in the x-y plane of the gas temperature at Lz/2 and the iron particles in the region Lz/2 ± 2*10^-4 m.
Left: Monodisperse. Right: Polydisperse.

We extended the previous monodisperse iron particle cloud case to a polydisperse distribution [2] by considering a realistic particle size range based on experiments from the EBI-VBT Bunsen flame.

In the monodisperse case (left), the particles uniformly ignite in a short time period, release heat through iron oxidation and cause a noticeable rise in gas temperature. In contrast, in the polydisperse case (right), smaller particles ignite first, but the associated heat release is not sufficient to significantly increase the surrounding gas. A more considerable temperature rise occurs only after larger particles ignite. As a net result, the bulk ignition in the polydisperse case is delayed, which is contrary to initial expectations. The simulation demonstrates that the particle size distribution has a crucial effect on the mixing process, ignition time and overall combustion behaviour. These effects should be considered in the design and optimisation of future iron-based particle combustors.

[1] Luu, T.D. et al. "Carrier-Phase DNS of Ignition and Combustion of Iron Particles in a Turbulent Mixing Layer". Flow Turbulence Combust 112 (2024) 1083-1103. 
⇒ https://doi.org/10.1007/s10494-023-00526-y
[2] Luu, T.D. et al. "Carrier-Phase DNS study of Particle Size Distribution Effects on Iron Particle Ignition in a Turbulent Mixing Layer". Proceedings of the Combustion Institute 40 (2024) 105297.
⇒ https://doi.org/10.1016/j.proci.2024.105297
[3] Thäter, G. et al. "The influence of clustering in homogeneous isotropic turbulence on the ignition behavior of iron particles". Proceedings of the Combustion Institute 40 (2024) 105348.
⇒ https://doi.org/10.1016/j.proci.2024.105348