Fast Computation of High Energy Elastic Collision Scattering Angle for Electric Propulsion Plume Simulation

In the plumes of Hall thrusters and ion thrusters, high energy ions experience elastic collisions with slow neutral atoms. These collisions involve a process of momentum exchange, altering the initial velocity vectors of the collision pair. In addition to the momentum exchange process, ions and atoms can exchange an electron, resulting in slow charge-exchange ions and fast atoms. In these simulations, it is particularly important to accurately perform computations of ion-atom elastic collisions in determining the plume current profile and assessing the integration of spacecraft components. The existing models are currently capable of accurate calculation but are not fast enough such that the calculation can be a bottleneck of plume simulations. This study investigates methods to accelerate an ion-atom elastic collision calculation that includes both momentumand charge-exchange processes. The scattering angles are pre-computed through a classical approach with ab initio spin-orbit free potential and are stored in a two-dimensional array as functions of impact parameter and energy. When performing a collision calculation for an ion-atom pair, the scattering angle is computed by a table lookup and multiple linear interpolations, given the relative energy and randomly determined impact parameter. In order to further accelerate the calculations, the number of collision calculations is reduced by properly defining an effective elastic collision cross-section that is used to decide if the degree of momentum exchange is insignificant. In the MCC method, the target atom needs to be sampled; however, it is confirmed that initial target atom velocity does not play significant role in typical electric propulsion plume simulations such that the sampling process is unnecessary. With these implementations, the computational run-time to perform a collision calculation is reduced significantly compared to previous methods, while retaining accuracy of the high fidelity models.