HiMomPlas · High-order moment models and multi-scale numerical methods for plasmas applied to electric propulsion

We are in the midst of a technological revolution characterized by the large demand for small satellites. These satellites are cheaper to launch and allow for forming constellations, impossible with larger satellites. As a result, they require propulsion systems in order to control their trajectories. Electric propulsion (EP) largely reduces the operational costs, using less propellant, precisely controlling thrust, and operating for long times. The response to the demand for EP is currently at stake because of two reasons: 1) Most of the thrusters operate with an expensive and scarce propellant gas, xenon, and 2) The development of the engines follows an empirical approach that is long and expensive, which limits the optimization and conception of new prototypes. These limitations are due to the lack of models that capture the complex plasma physics in EP, in particular: 1) The impact of the plasma composition and thruster geometry on the performance, 2) The presence of multi-scale plasma kinetic instabilities, and 3) The erosion of the walls interacting with the plasma. Our objective is to develop advanced predictive plasma models, able to help to the conception of new thrusters operating with cheaper propellants, improving their stability and performance, and predicting their lifetime, without the need of expensive experimental campaigns. We will follow a multidisciplinary approach, integrating physicist, engineers, chemists, and applied mathematicians. Our methodology aims at extending the classical fluid equations by solving for higher-order moments, allowing to describe in an efficient manner microscopic (kinetic) processes. With advanced numerical techniques, we will propose multi-scale models able to describe the turbulent phenomena, the wall processes, and efficiently simulate the multi-scale nature of EP. The models will be integrated into a numerical tool that will be validated with experiments using cutting-edge uncertainty quantification techniques.