Direct simulation Monte Carlo for modeling spatially homogeneous multi-component aerosols

Abstract

Nuclear aerosols generated under normal operational and post-accident reactor environments are of particular importance in estimation of the nuclear source term. Several light water reactor aerosol containment experiments provided an experimental database for verification and validation of thermal-hydraulic and aerosol transport codes. The Direct Simulation Monte Carlo (DSMC) technique has been shown to model multicomponent aerosol dynamics accurately while maintaining greater fidelity to actual aerosol physics than its sectional, moments, and finite element predecessors. This research focuses on the development of a comprehensive n-component source term code for modeling the behavior of aerosolized fission products based on the DSMC technique. Effective DSMC benchmarks provided further confidence in the technique's capabilities for modeling exceedingly complex systems. With the inclusion of the Knudsen, Kelvin, and solute effects in the Mason model, the role of condensation on aerosol evolution showed the differentiation of particles by physical size and chemical properties. High fidelity large-scale simulations posed evident but considerable challenges to computational runtime. Developments in the simulation scaling theory for coagulation, condensation, deposition, and generation processes showed to give comparable results while simultaneously reducing simulation time significantly. The evolution of aerosols coupled to environments was explored, and benchmark simulations provided further evidence that DSMC accurately models aerosol dynamics when coupled with containment thermal-hydraulics.