Numerical simulation of aerosol dynamics in multi-scale systems

Abstract

The study of aerosol dynamics is of great importance to a variety of scientific fields including air pollution, vehicle emissions, combustion and chemical engineering science. A new stochastically weighted operator splitting Monte Carlo (SWOSMC) method is first proposed and developed in the present study in which weighted numerical particles and operator splitting technique are coupled in order to reduce statistical error and accelerate the simulation of particle-fluid systems undergoing simultaneous complex aerosol dynamic processes. This new SWOSMC method is first validated by comparing its simulation results with the corresponding analytical solution for the selected cases. Some cases involving the evolution and formation of complex particle processes in fluid-particle systems are studied using this SWOSMC method. The obtained results are compared with those obtained by the sectional method and good agreement is obtained. Computational analysis indicates that this new SWOSMC method has high computational efficiency and accuracy in solving complex particle-fluid system problems, particularly simultaneous aerosol dynamic processes. In order to solve multi-dimensional aerosols dynamics interacting with continuous fluid phase, this validated SWOSMC method for population balance equation (PBE) is coupled with computational fluid dynamics (CFD) under the Eulerian-Lagrangian reference frame. The formulated CFD-Monte Carlo (CFD-MC) method is used to study complex aerosol dynamics in turbulent flows. Several typical cases of aerosol dynamic processes including turbulent coagulation, nucleation and growth are studied and compared to the population balance sectional method (PBSM) with excellent agreement. The effects of different jet Reynolds (Rej) numbers on aerosol dynamics in turbulent flows are fully investigated for each of the studied cases in an aerosol reactor. The results demonstrate that Rej has significant impact on a single aerosol dynamic process (e.g. coagulation) as well as the competition between simultaneous aerosol dynamic processes in turbulent flows. This newly proposed and developed CFD-Monte Carlo/probability density function (CFD-MC/PDF) method renders an efficient method for simulating complex aerosol dynamics in turbulent flows and provides a better insight into the interaction between turbulence and the full particle size distribution (PSD) of aerosol particles. Finally, aerosol dynamics in turbulent reactive flows i.e., soot dynamics in turbulent reactive flows, is investigated and validated with corresponding experimental results available in literature. Excellent numerical results in temperature, mixture fraction and soot volume fraction as well as PSD of soot particles are obtained when compared with the experimental results, which validates the capability of this new CFD-MC/PDF method with the soot and radiation models for solving aerosol dynamics in turbulent reactive flows. In summary, this newly proposed and developed CFD-MC/PDF method in the present study has demonstrated high capability, and computational efficiency and accuracy in the numerical simulation of complex aerosol dynamics in multi-scale systems.