A Theory for Close-Coupled Gas Atomization Process Revisited: A Visual, Computational and Empirical Study

摘要

A review of an atomization model is represented here to explain the pulsatile atomization theory in liquid metal atomization for improved control of the powder production process. The model takes the basic observations from the fuel combustion, in aerospace applications, and applies it to gas atomization of molten metal for metal powder production. The common premise of both processes is in the utilizing of the high pressure gas kinetic energy to overcome the surface tension of the liquid fluid disintegrating the liquid into droplets. Understanding the gas dynamic and liquid metal flow rate behaviors at the melt orifice provided a premise for further understanding of the atomization process towards a cost efficient production of fine metal powder for the additive manufacturing and MIM powder space. Although the gas and melt interaction appears to be chaotic to the naked-eye, these parameters are in fact interdependent, thus possibly enabling one to control the gas atomization process. In the past, it has been commonly reported that the finest metal particle production has been found when operating an atomizer at strong subambient pressure at the melt orifice. This subambient pressure created by a gas dynamic phenomenon known as closed-wake condition has also been commonly associated with robust metal liquid (melt) atomization enhanced by, supposedly, increased melt flow rate. These intuitive assumptions, however, has been shown to be in fact erroneous. Contrary to prevailing beliefs, it has been proposed in a series of studies that the melt flow rate actually retards when operating the atomizer at pressures that causes the closed-wake condition. As a matter of fact, contrary to intuitive assumptions, deep subambient pressure at the melt orifice retards the melt flows during atomization. Once again, the theory of this two-phase fluid phenomenon in melt gas atomization is discussed in this pulsatile atomization model. This model is computationally supported by a CFD modeling, and further validated via a high-speed video recording of a melt gas atomization process. All this will be discussed in this paper. The empirical study of the atomization model was first done on an industrial melt atomizer unit using molten metal charges of a nickel-based superalloy. A pulsatile atomization model is presented to bridge the relationship between gas-only aspiration pressure and the resulting particle yield from the melt atomization. The CFD modeling was performed on a similar closed-coupled nozzle configuration. Finally the high speed video recording was performed on a similar nozzle configuration.