A hybrid computational framework for the simulation of atmospheric pressure plasma jets: the importance of the gas flow model

Abstract

A hybrid computational framework, consisting of a detailed turbulence flow model, a global model, and a model for the calculation of the electron energy probability function, is developed to predict the density of plasma generated species along the axial direction of plasma jets. The framework is applied to an Ar/O2 plasma in a kINPen 09 device without a shielding gas. A reaction set of 764 reactions and 84 species is considered. The effect of different turbulence flow models, namely the detailed and high cost large eddy simulation (LES) model and the simple and low cost realizable k–ε model, on the densities of plasma generated species is investigated at different values of absorbed power. The effect is not severe on the density of the majority of the species, justified by the small differences in the inputs of the global model, i.e. the volume averaged axial velocity and density of air species (coming from the turbulence flow model). Nevertheless, the differences in the densities of O2(1Σg), O−, O2 −, O(1D), O, H, H2(r), H−, N2O(v), H7O3 +, H9O4 +, H15O7 + and OH− are remarkably affected by the choice of the turbulence flow model and may reach an order of magnitude. The detailed LES model is a proper choice for Ar jets and this is reinforced by the comparison of the results of the framework with atomic oxygen experimental measurements along the axial direction of the jet: the use of the LES model leads to atomic oxygen density closer to the measured one compared to (the use of) the realizable k–ε model. Finally, an evaluation of the assumptions required for the use of global models in plasma jets is performed, demonstrating their validity for the case studied.