Gas atomisation is the process of producing melt powders for additive manufacturing applications. The process involves high-speed gas flow to break-up and disintegrate molten melt into ultra-fine metal powders about the order of micrometers (μm). Although gas atomisation is the commercially viable process to produce realisable yield to meet the industrial scale production, yet the process is highly inefficient due to its chaotic and unpredictable gas-melt interaction, resulting in increased particle size distributions.
This thesis presents a series of 22 computational fluid dynamics (CFD) simulations, numerically modelling close-coupled gas atomisation (CCGA) process to investigate the gas-melt interactions and their individual instabilities under varying gas-to-melt ratios (GMRs), and melt temperatures. The premise of these simulations is to understand the coupled nature of the gas-melt interactions resulting in a chaotic and unpredictable process, as typically seen in the physical atomisation process. To capture the gas-melt interactions using a CFD model, the melt flow is coupled to the atomiser’s internal pressure (aspiration pressure), which varies over time due to the unsteady behaviour of the gas flow-field, induced by the gas-melt interactions. The CFD simulations are modelled as two-phase flow, with Argon gas as the primary phase, and melt as the secondary phase (particles), modelled via coupled Euler-Lagrange framework. The CFD study was investigated using five different GMRs and melt temperatures: GMRs of 5.5, 2.6, 1.32, 0.88, and 0.44, and melt temperature of 300K, 600K, 1000K, 1500K, and 2000K, and they were run in various combinations. These series of simulations investigated into the gas and melt instabilities and their association with each other.
The study, via CFD models revealed that the atomisation process at higher GMR (5.5 & 2.6) is extremely stable due to the dominance of the gas flow, and that it effectively attenuates any external instability induced into the system. Similarly, investigation into varying GMRs and melt temperatures exhibited that with decreasing GMR or increasing melt temperature, a melt instability was introduced into the process. The melt instability observed across decreasing GMR or increasing melt temperature were profound and displayed multi-layered frequency of fluctuations. However, the effect of melt temperature on the process is more prominent and substantial compared to that of the GMR. This is manifested by the evolving complexity and unpredictability in the process with increase in melt temperature.
