Computational fluid dynamics predicts the nanoparticle transport in gas aggregation cluster sources

Abstract

In a typical sputter-based gas aggregation cluster source (GAS), nanoparticles (NPs) are created from supersaturated vapours of the target material. The NPs then escape from the source with the expanding gas through an exit orifice. The carrier gas flow profile is one of the most critical parameters, which strongly affects the NP losses on the walls and determines the efficiency of the NP transport to the substrate. In this work, computational fluid dynamics (CFD) simulations are performed to understand the flow of the carrier gas inside the aggregation chamber. We focus on the impact of the inlet and outlet geometry on the carrier gas flow and, therefore, on the NP transportation. Two types of GAS with either a conventional planar magnetron or a cylindrical magnetron are considered. In the planar configuration, the working gas inlet is from behind the magnetron, and the gas flows around the target towards the orifice along the system axis, which may cause some vertices. The situation is even more critical for the cylindrical magnetron, where the gas inlet position and geometry have a drastic influence on the gas flow. Brownian diffusion is found to prevail for NPs smaller than 5 nm, regardless of the gas flow. This leads to their losses on the walls. Larger NPs experience a stronger drag force from the carrier gas flow, which should exceed 10 m s−1 to prevent loss of NPs on the walls and keep NP transport efficient. Therefore, the CFD simulations help to visualise the motion of the NPs and optimise the geometry of the GAS for future applications.