Numerical investigation of the radiative properties of a fully resolved particle cloud

Abstract

In this study, the interaction of thermal radiation with dense solid fuel particle clouds is investigated numerically. To determine absorption and scattering properties of a particle cloud, generally, the absorption and scattering efficiencies of single particles are calculated by Mie theory, and subsequently, these parameters are linearly scaled by the particle number density to obtain representative coefficients for the particle cloud. Theoretically, this approach is only valid for clouds with perfectly uniformly distributed particles. However, real particle clouds are characterized by non-uniform particle distributions within the cloud. This results in the radiation being able to penetrate areas with low particle concentration almost unhindered, while areas with high particle concentrations are subject to multiple particle-radiation interactions.To evaluate the impact of non-uniform particle distributions on the radiative heat transfer, generic particle clouds with random distributions are generated. The cell size of the numerical grid is chosen in such a way that each individual particle is resolved, while each cell either represents particle or gas properties. The radiative heat transfer through the particle cloud is solved numerically via the Monte-Carlo ray-tracing method. The particle-resolved approach and the linear scaled approach are compared based on the intensity loss as a function of penetrating depth.The absorption behavior of a particle cloud modeled by a non-uniform particle distribution indicates a value up to 30% lower than the one of a perfectly uniform particle distribution. The investigations have further shown that it is possible to consider the effects of the particle-resolved approach by a modified weighting factor, which takes into account the particle size and absorption efficiency in addition to the particle number density. This allows numerical calculations without resolution of individual particles (low numerical effort) with similar accuracy compared to the particle-resolved simulations. Since dependent scattering effects are not modeled, the modified weighting factor is only applicable for particle clouds with a particle volume fraction below P<0.01.