Challenges with the self-consistent implementation of closure models for anomalous electron transport in fluid simulations of Hall thrusters

The performance of closure models for the anomalous electron transport when self-consistently implemented in a fluid model for a Hall effect thruster is investigated. This cross-field transport, which is orders of magnitude higher than classical collisional transport, is represented as an effective collision frequency. The proposed closure models relate this transport coefficient to local fluid properties of the plasma. Before implementation, the models are calibrated against values of the collision frequency inferred empirically from a 9 kW Hall thruster at 300 V and 15 A. It is found that even though closure models match the empirical collision frequency values, they diverge from these values when implemented self-consistently in a Hall thruster code. Possible drivers of this behavior are examined, including the role of non-linearity in the governing equations of the Hall thruster fluid model, artifacts from using time-averaged calibration data, and the non-uniqueness of the empirically-inferred collision frequencies. These results are discussed in the context of their implications for discovering and validating new closures necessary for enabling fully-predictive Hall thruster models.